20200616.md

ArXiv cs.CV --Tue, 16 Jun 2020

1.Coherent Reconstruction of Multiple Humans from a Single Image ⬇️

In this work, we address the problem of multi-person 3D pose estimation from a single image. A typical regression approach in the top-down setting of this problem would first detect all humans and then reconstruct each one of them independently. However, this type of prediction suffers from incoherent results, e.g., interpenetration and inconsistent depth ordering between the people in the scene. Our goal is to train a single network that learns to avoid these problems and generate a coherent 3D reconstruction of all the humans in the scene. To this end, a key design choice is the incorporation of the SMPL parametric body model in our top-down framework, which enables the use of two novel losses. First, a distance field-based collision loss penalizes interpenetration among the reconstructed people. Second, a depth ordering-aware loss reasons about occlusions and promotes a depth ordering of people that leads to a rendering which is consistent with the annotated instance segmentation. This provides depth supervision signals to the network, even if the image has no explicit 3D annotations. The experiments show that our approach outperforms previous methods on standard 3D pose benchmarks, while our proposed losses enable more coherent reconstruction in natural images. The project website with videos, results, and code can be found at: this https URL

2.Now that I can see, I can improve: Enabling data-driven finetuning of CNNs on the edge ⬇️

In today's world, a vast amount of data is being generated by edge devices that can be used as valuable training data to improve the performance of machine learning algorithms in terms of the achieved accuracy or to reduce the compute requirements of the model. However, due to user data privacy concerns as well as storage and communication bandwidth limitations, this data cannot be moved from the device to the data centre for further improvement of the model and subsequent deployment. As such there is a need for increased edge intelligence, where the deployed models can be fine-tuned on the edge, leading to improved accuracy and/or reducing the model's workload as well as its memory and power footprint. In the case of Convolutional Neural Networks (CNNs), both the weights of the network as well as its topology can be tuned to adapt to the data that it processes. This paper provides a first step towards enabling CNN finetuning on an edge device based on structured pruning. It explores the performance gains and costs of doing so and presents an extensible open-source framework that allows the deployment of such approaches on a wide range of network architectures and devices. The results show that on average, data-aware pruning with retraining can provide 10.2pp increased accuracy over a wide range of subsets, networks and pruning levels with a maximum improvement of 42.0pp over pruning and retraining in a manner agnostic to the data being processed by the network.

3.Visibility Guided NMS: Efficient Boosting of Amodal Object Detection in Crowded Traffic Scenes ⬇️

Object detection is an important task in environment perception for autonomous driving. Modern 2D object detection frameworks such as Yolo, SSD or Faster R-CNN predict multiple bounding boxes per object that are refined using Non-Maximum-Suppression (NMS) to suppress all but one bounding box. While object detection itself is fully end-to-end learnable and does not require any manual parameter selection, standard NMS is parametrized by an overlap threshold that has to be chosen by hand. In practice, this often leads to an inability of standard NMS strategies to distinguish different objects in crowded scenes in the presence of high mutual occlusion, e.g. for parked cars or crowds of pedestrians. Our novel Visibility Guided NMS (vg-NMS) leverages both pixel-based as well as amodal object detection paradigms and improves the detection performance especially for highly occluded objects with little computational overhead. We evaluate vg-NMS using KITTI, VIPER as well as the Synscapes dataset and show that it outperforms current state-of-the-art NMS.

4.Go-CaRD -- Generic, Optical Car Part Recognition and Detection: Collection, Insights, and Applications ⬇️

Systems for the automatic recognition and detection of automotive parts are crucial in several emerging research areas in the development of intelligent vehicles. They enable, for example, the detection and modelling of interactions between human and the vehicle. In this paper, we present three suitable datasets as well as quantitatively and qualitatively explore the efficacy of state-of-the-art deep learning architectures for the localisation of 29 interior and exterior vehicle regions, independent of brand, model, and environment. A ResNet50 model achieved an F1 score of 93.67 % for recognition, while our best Darknet model achieved an mAP of 58.20 % for detection. We also experiment with joint and transfer learning approaches and point out potential applications of our systems.

5.Towards Incorporating Contextual Knowledge into the Prediction of Driving Behavior ⬇️

Predicting the behavior of surrounding traffic participants is crucial for advanced driver assistance systems and autonomous driving. Most researchers however do not consider contextual knowledge when predicting vehicle motion. Extending former studies, we investigate how predictions are affected by external conditions. To do so, we categorize different kinds of contextual information and provide a carefully chosen definition as well as examples for external conditions. More precisely, we investigate how a state-of-the-art approach for lateral motion prediction is influenced by one selected external condition, namely the traffic density. Our investigations demonstrate that this kind of information is highly relevant in order to improve the performance of prediction algorithms. Therefore, this study constitutes the first step towards the integration of such information into automated vehicles. Moreover, our motion prediction approach is evaluated based on the public highD data set showing a maneuver prediction performance with areas under the ROC curve above 97% and a median lateral prediction error of only 0.18m on a prediction horizon of 5s.

6.3D-ZeF: A 3D Zebrafish Tracking Benchmark Dataset ⬇️

In this work we present a novel publicly available stereo based 3D RGB dataset for multi-object zebrafish tracking, called 3D-ZeF. Zebrafish is an increasingly popular model organism used for studying neurological disorders, drug addiction, and more. Behavioral analysis is often a critical part of such research. However, visual similarity, occlusion, and erratic movement of the zebrafish makes robust 3D tracking a challenging and unsolved problem. The proposed dataset consists of eight sequences with a duration between 15-120 seconds and 1-10 free moving zebrafish. The videos have been annotated with a total of 86,400 points and bounding boxes. Furthermore, we present a complexity score and a novel open-source modular baseline system for 3D tracking of zebrafish. The performance of the system is measured with respect to two detectors: a naive approach and a Faster R-CNN based fish head detector. The system reaches a MOTA of up to 77.6%. Links to the code and dataset is available at the project page this https URL

7.SD-RSIC: Summarization Driven Deep Remote Sensing Image Captioning ⬇️

Deep neural networks (DNNs) have been recently found popular for image captioning problems in remote sensing (RS). Existing DNN based approaches rely on the availability of a training set made up of a high number of RS images with their captions. However, captions of training images may contain redundant information (they can be repetitive or semantically similar to each other), resulting in information deficiency while learning a mapping from image domain to language domain. To overcome this limitation, in this paper we present a novel Summarization Driven Remote Sensing Image Captioning (SD-RSIC) approach. The proposed approach consists of three main steps. The first step obtains the standard image captions by jointly exploiting convolutional neural networks (CNNs) with long short-term memory (LSTM) networks. The second step, unlike the existing RS image captioning methods, summarizes the ground-truth captions of each training image into a single caption by exploiting sequence to sequence neural networks and eliminates the redundancy present in the training set. The third step automatically defines the adaptive weights associated to each RS image to combine the standard captions with the summarized captions based on the semantic content of the image. This is achieved by a novel adaptive weighting strategy defined in the context of LSTM networks. Experimental results obtained on the RSCID, UCM-Captions and Sydney-Captions datasets show the effectiveness of the proposed approach compared to the state-of-the-art RS image captioning approaches.

8.Pixel Invisibility: Detecting Objects Invisible in Color Images ⬇️

Despite recent success of object detectors using deep neural networks, their deployment on safety-critical applications such as self-driving cars remains questionable. This is partly due to the absence of reliable estimation for detectors' failure under operational conditions such as night, fog, dusk, dawn and glare. Such unquantifiable failures could lead to safety violations. In order to solve this problem, we created an algorithm that predicts a pixel-level invisibility map for color images that does not require manual labeling - that computes the probability that a pixel/region contains objects that are invisible in color domain, during various lighting conditions such as day, night and fog. We propose a novel use of cross modal knowledge distillation from color to infra-red domain using weakly-aligned image pairs from the day and construct indicators for the pixel-level invisibility based on the distances of their intermediate-level features. Quantitative experiments show the great performance of our pixel-level invisibility mask and also the effectiveness of distilled mid-level features on object detection in infra-red imagery.

9.Generating Master Faces for Use in Performing Wolf Attacks on Face Recognition Systems ⬇️

Due to its convenience, biometric authentication, especial face authentication, has become increasingly mainstream and thus is now a prime target for attackers. Presentation attacks and face morphing are typical types of attack. Previous research has shown that finger-vein- and fingerprint-based authentication methods are susceptible to wolf attacks, in which a wolf sample matches many enrolled user templates. In this work, we demonstrated that wolf (generic) faces, which we call "master faces," can also compromise face recognition systems and that the master face concept can be generalized in some cases. Motivated by recent similar work in the fingerprint domain, we generated high-quality master faces by using the state-of-the-art face generator StyleGAN in a process called latent variable evolution. Experiments demonstrated that even attackers with limited resources using only pre-trained models available on the Internet can initiate master face attacks. The results, in addition to demonstrating performance from the attacker's point of view, can also be used to clarify and improve the performance of face recognition systems and harden face authentication systems.

10.Tamil Vowel Recognition With Augmented MNIST-like Data Set ⬇️

We report generation of a MNIST [4] compatible data set [1] for Tamil vowels to enable building a classification DNN or other such ML/AI deep learning [2] models for Tamil OCR/Handwriting applications. We report the capability of the 60,000 grayscale, 28x28 pixel dataset to build a 92% accuracy (training) and 82% cross-validation 4-layer CNN, with 100,000+ parameters, in TensorFlow. We also report a top-1 classification accuracy of 70% and top-2 classification accuracy of 92% on handwritten vowels showing, for the same network.

11.CoDeNet: Algorithm-hardware Co-design for Deformable Convolution ⬇️

Deploying deep learning models on embedded systems for computer vision tasks has been challenging due to limited compute resources and strict energy budgets. The majority of existing work focuses on accelerating image classification, while other fundamental vision problems, such as object detection, have not been adequately addressed. Compared with image classification, detection problems are more sensitive to the spatial variance of objects, and therefore, require specialized convolutions to aggregate spatial information. To address this, recent work proposes dynamic deformable convolution to augment regular convolutions. Regular convolutions process a fixed grid of pixels across all the spatial locations in an image, while dynamic deformable convolution may access arbitrary pixels in the image and the access pattern is input-dependent and varies per spatial location. These properties lead to inefficient memory accesses of inputs with existing hardware. In this work, we first investigate the overhead of the deformable convolution on embedded FPGA SoCs, and introduce a depthwise deformable convolution to reduce the total number of operations required. We then show the speed-accuracy tradeoffs for a set of algorithm modifications including irregular-access versus limited-range and fixed-shape. We evaluate these algorithmic changes with corresponding hardware optimizations. Results show a 1.36x and 9.76x speedup respectively for the full and depthwise deformable convolution on the embedded FPGA accelerator with minor accuracy loss on the object detection task. We then co-design an efficient network CoDeNet with the modified deformable convolution for object detection and quantize the network to 4-bit weights and 8-bit activations. Results show that our designs lie on the pareto-optimal front of the latency-accuracy tradeoff for the object detection task on embedded FPGAs

12.ORD: Object Relationship Discovery for Visual Dialogue Generation ⬇️

With the rapid advancement of image captioning and visual question answering at single-round level, the question of how to generate multi-round dialogue about visual content has not yet been well explored.Existing visual dialogue methods encode the image into a fixed feature vector directly, concatenated with the question and history embeddings to predict the response.Some recent methods tackle the co-reference resolution problem using co-attention mechanism to cross-refer relevant elements from the image, history, and the target question.However, it remains challenging to reason visual relationships, since the fine-grained object-level information is omitted before co-attentive reasoning. In this paper, we propose an object relationship discovery (ORD) framework to preserve the object interactions for visual dialogue generation. Specifically, a hierarchical graph convolutional network (HierGCN) is proposed to retain the object nodes and neighbour relationships locally, and then refines the object-object connections globally to obtain the final graph embeddings. A graph attention is further incorporated to dynamically attend to this graph-structured representation at the response reasoning stage. Extensive experiments have proved that the proposed method can significantly improve the quality of dialogue by utilising the contextual information of visual relationships. The model achieves superior performance over the state-of-the-art methods on the Visual Dialog dataset, increasing MRR from 0.6222 to 0.6447, and recall@1 from 48.48% to 51.22%.

13.On the Preservation of Spatio-temporal Information in Machine Learning Applications ⬇️

In conventional machine learning applications, each data attribute is assumed to be orthogonal to others. Namely, every pair of dimension is orthogonal to each other and thus there is no distinction of in-between relations of dimensions. However, this is certainly not the case in real world signals which naturally originate from a spatio-temporal configuration. As a result, the conventional vectorization process disrupts all of the spatio-temporal information about the order/place of data whether it be $1$D, $2$D, $3$D, or $4$D. In this paper, the problem of orthogonality is first investigated through conventional $k$-means of images, where images are to be processed as vectors. As a solution, shift-invariant $k$-means is proposed in a novel framework with the help of sparse representations. A generalization of shift-invariant $k$-means, convolutional dictionary learning, is then utilized as an unsupervised feature extraction method for classification. Experiments suggest that Gabor feature extraction as a simulation of shallow convolutional neural networks provides a little better performance compared to convolutional dictionary learning. Many alternatives of convolutional-logic are also discussed for spatio-temporal information preservation, including a spatio-temporal hypercomplex encoding scheme.

14.Mitigating Gender Bias in Captioning Systems ⬇️

Recent studies have shown that captioning datasets, such as the COCO dataset, may contain severe social bias which could potentially lead to unintentional discrimination in learning models. In this work, we specifically focus on the gender bias problem. The existing dataset fails to quantify bias because models that intrinsically memorize gender bias from training data could still achieve a competitive performance on the biased test dataset. To bridge the gap, we create two new splits: COCO-GB v1 and v2 to quantify the inherent gender bias which could be learned by models. Several widely used baselines are evaluated on our new settings, and experimental results indicate that most models learn gender bias from the training data, leading to an undesirable gender prediction error towards women. To overcome the unwanted bias, we propose a novel Guided Attention Image Captioning model (GAIC) which provides self-guidance on visual attention to encourage the model to explore correct gender visual evidence. Experimental results validate that GAIC can significantly reduce gender prediction error, with a competitive caption quality. Our codes and the designed benchmark datasets are available at this https URL.

15.Deep-CAPTCHA: a deep learning based CAPTCHA solver for vulnerability assessment ⬇️

CAPTCHA is a human-centred test to distinguish a human operator from bots, attacking programs, or any other computerised agent that tries to imitate human intelligence. In this research, we investigate a way to crack visual CAPTCHA tests by an automated deep learning based solution. The goal of the cracking is to investigate the weaknesses and vulnerabilities of the CAPTCHA generators and to develop more robust CAPTCHAs, without taking the risks of manual try and error efforts. We have developed a Convolutional Neural Network called \Deep-CAPTCHA to achieve this goal. We propose a platform to investigate both numerical and alphanumerical image CAPTCHAs. To train and develop an efficient model, we have generated 500,000 CAPTCHAs using Python Image-Captcha Library. In this paper, we present our customised deep neural network model, the research gaps and the existing challenges, and the solutions to overcome the issues. Our network's cracking accuracy results leads to 98.94% and 98.31% for the numerical and the alpha-numerical Test datasets, respectively. That means more works need to be done to develop robust CAPTCHAs, to be non-crackable against bot attaches and artificial agents. As the outcome of this research, we identify some efficient techniques to improve the CAPTCHA generators, based on the performance analysis conducted on the Deep-CAPTCHA model.

16.AMENet: Attentive Maps Encoder Network for Trajectory Prediction ⬇️

Trajectory prediction is a crucial task in different communities, such as intelligent transportation systems, photogrammetry, computer vision, and mobile robot applications. However, there are many challenges to predict the trajectories of heterogeneous road agents (e.g. pedestrians, cyclists and vehicles) at a microscopical level. For example, an agent might be able to choose multiple plausible paths in complex interactions with other agents in varying environments, and the behavior of each agent is affected by the various behaviors of its neighboring agents. To this end, we propose an end-to-end generative model named Attentive Maps Encoder Network (AMENet) for accurate and realistic multi-path trajectory prediction. Our method leverages the target road user's motion information (i.e. movement in xy-axis in a Cartesian space) and the interaction information with the neighboring road users at each time step, which is encoded as dynamic maps that are centralized on the target road user. A conditional variational auto-encoder module is trained to learn the latent space of possible future paths based on the dynamic maps and then used to predict multiple plausible future trajectories conditioned on the observed past trajectories. Our method reports the new state-of-the-art performance (final/mean average displacement (FDE/MDE) errors 1.183/0.356 meters) on benchmark datasets and wins the first place in the open challenge of Trajnet.

17.Dermatologist vs Neural Network ⬇️

Cancer, in general, is very deadly. Timely treatment of any cancer is the key to saving a life. Skin cancer is no exception. There have been thousands of Skin Cancer cases registered per year all over the world. There have been 123,000 deadly melanoma cases detected in a single year. This huge number is proven to be a cause of a high amount of UV rays present in the sunlight due to the degradation of the Ozone layer. If not detected at an early stage, skin cancer can lead to the death of the patient. Unavailability of proper resources such as expert dermatologists, state of the art testing facilities, and quick biopsy results have led researchers to develop a technology that can solve the above problem. Deep Learning is one such method that has offered extraordinary results. The Convolutional Neural Network proposed in this study out performs every pretrained models. We trained our model on the HAM10000 dataset which offers 10015 images belonging to 7 classes of skin disease. The model we proposed gave an accuracy of 89%. This model can predict deadly melanoma skin cancer with a great accuracy. Hopefully, this study can help save people's life where there is the unavailability of proper dermatological resources by bridging the gap using our proposed study.

18.Learn to cycle: Time-consistent feature discovery for action recognition ⬇️

Temporal motion has been one of the essential components for effectively recognizing actions in videos. Both, time information and features are primarily extracted hierarchically through small sequences of few frames, with the use of 3D convolutions. In this paper, we propose a method that can learn general feature changes across time, making activations unbounded to a temporal locality, by additionally including a general notion of their learned features. Through this recalibration of temporal feature cues across multiple frames, 3D-CNN models are capable of using features that are prevalent over different time segments, while being less constraint by their temporal receptive fields. We present improvements on both high and low capacity models, with the largest benefits being observed in low-memory models, as most of their current drawbacks rely on their poor generalization capabilities because of the low number and feature complexity. We present average improvements, over both corresponding and state-of-the-art models, in the range of 3.67% on Kinetics-700 (K-700), 2.75% on Moments in Time (MiT), 2.57% on Human Actions Clips and Segments (HACS), 3.195% on HMDB-51 and 3.30% on UCF-101.

19.AutoGAN-Distiller: Searching to Compress Generative Adversarial Networks ⬇️

The compression of Generative Adversarial Networks (GANs) has lately drawn attention, due to the increasing demand for deploying GANs into mobile devices for numerous applications such as image translation, enhancement and editing. However, compared to the substantial efforts to compressing other deep models, the research on compressing GANs (usually the generators) remains at its infancy stage. Existing GAN compression algorithms are limited to handling specific GAN architectures and losses. Inspired by the recent success of AutoML in deep compression, we introduce AutoML to GAN compression and develop an AutoGAN-Distiller (AGD) framework. Starting with a specifically designed efficient search space, AGD performs an end-to-end discovery for new efficient generators, given the target computational resource constraints. The search is guided by the original GAN model via knowledge distillation, therefore fulfilling the compression. AGD is fully automatic, standalone (i.e., needing no trained discriminators), and generically applicable to various GAN models. We evaluate AGD in two representative GAN tasks: image translation and super resolution. Without bells and whistles, AGD yields remarkably lightweight yet more competitive compressed models, that largely outperform existing alternatives.

20.Infinite Feature Selection: A Graph-based Feature Filtering Approach ⬇️

We propose a filtering feature selection framework that considers subsets of features as paths in a graph, where a node is a feature and an edge indicates pairwise (customizable) relations among features, dealing with relevance and redundancy principles. By two different interpretations (exploiting properties of power series of matrices and relying on Markov chains fundamentals) we can evaluate the values of paths (i.e., feature subsets) of arbitrary lengths, eventually go to infinite, from which we dub our framework Infinite Feature Selection (Inf-FS). Going to infinite allows to constrain the computational complexity of the selection process, and to rank the features in an elegant way, that is, considering the value of any path (subset) containing a particular feature. We also propose a simple unsupervised strategy to cut the ranking, so providing the subset of features to keep. In the experiments, we analyze diverse settings with heterogeneous features, for a total of 11 benchmarks, comparing against 18 widely-known comparative approaches. The results show that Inf-FS behaves better in almost any situation, that is, when the number of features to keep are fixed a priori, or when the decision of the subset cardinality is part of the process.

21.Binary DAD-Net: Binarized Driveable Area Detection Network for Autonomous Driving ⬇️

Driveable area detection is a key component for various applications in the field of autonomous driving (AD), such as ground-plane detection, obstacle detection and maneuver planning. Additionally, bulky and over-parameterized networks can be easily forgone and replaced with smaller networks for faster inference on embedded systems. The driveable area detection, posed as a two class segmentation task, can be efficiently modeled with slim binary networks. This paper proposes a novel binarized driveable area detection network (binary DAD-Net), which uses only binary weights and activations in the encoder, the bottleneck, and the decoder part. The latent space of the bottleneck is efficiently increased (x32 -> x16 downsampling) through binary dilated convolutions, learning more complex features. Along with automatically generated training data, the binary DAD-Net outperforms state-of-the-art semantic segmentation networks on public datasets. In comparison to a full-precision model, our approach has a x14.3 reduced compute complexity on an FPGA and it requires only 0.9MB memory resources. Therefore, commodity SIMD-based AD-hardware is capable of accelerating the binary DAD-Net.

22.Neural gradients are lognormally distributed: understanding sparse and quantized training ⬇️

Neural gradient compression remains a main bottleneck in improving training efficiency, as most existing neural network compression methods (e.g., pruning or quantization) focus on weights, activations, and weight gradients. However, these methods are not suitable for compressing neural gradients, which have a very different distribution. Specifically, we find that the neural gradients follow a lognormal distribution. Taking this into account, we suggest two methods to reduce the computational and memory burdens of neural gradients. The first one is stochastic gradient pruning, which can accurately set the sparsity level -- up to 85% gradient sparsity without hurting validation accuracy (ResNet18 on ImageNet). The second method determines the floating-point format for low numerical precision gradients (e.g., FP8). Our results shed light on previous findings related to local scaling, the optimal bit-allocation for the mantissa and exponent, and challenging workloads for which low-precision floating-point arithmetic has reported to fail. Reference implementation accompanies the paper.

23.Filter design for small target detection on infrared imagery using normalized-cross-correlation layer ⬇️

In this paper, we introduce a machine learning approach to the problem of infrared small target detection filter design. For this purpose, similarly to a convolutional layer of a neural network, the normalized-cross-correlational (NCC) layer, which we utilize for designing a target detection/recognition filter bank, is proposed. By employing the NCC layer in a neural network structure, we introduce a framework, in which supervised training is used to calculate the optimal filter shape and the optimum number of filters required for a specific target detection/recognition task on infrared images. We also propose the mean-absolute-deviation NCC (MAD-NCC) layer, an efficient implementation of the proposed NCC layer, designed especially for FPGA systems, in which square root operations are avoided for real-time computation. As a case study we work on dim-target detection on mid-wave infrared imagery and obtain the filters that can discriminate a dim target from various types of background clutter, specific to our operational concept.

24.Survey on Deep Multi-modal Data Analytics: Collaboration, Rivalry and Fusion ⬇️

With the development of web technology, multi-modal or multi-view data has surged as a major stream for big data, where each modal/view encodes individual property of data objects. Often, different modalities are complementary to each other. Such fact motivated a lot of research attention on fusing the multi-modal feature spaces to comprehensively characterize the data objects. Most of the existing state-of-the-art focused on how to fuse the energy or information from multi-modal spaces to deliver a superior performance over their counterparts with single modal. Recently, deep neural networks have exhibited as a powerful architecture to well capture the nonlinear distribution of high-dimensional multimedia data, so naturally does for multi-modal data. Substantial empirical studies are carried out to demonstrate its advantages that are benefited from deep multi-modal methods, which can essentially deepen the fusion from multi-modal deep feature spaces. In this paper, we provide a substantial overview of the existing state-of-the-arts on the filed of multi-modal data analytics from shallow to deep spaces. Throughout this survey, we further indicate that the critical components for this field go to collaboration, adversarial competition and fusion over multi-modal spaces. Finally, we share our viewpoints regarding some future directions on this field.

25.Classifying degraded images over various levels of degradation ⬇️

Classification for degraded images having various levels of degradation is very important in practical applications. This paper proposes a convolutional neural network to classify degraded images by using a restoration network and an ensemble learning. The results demonstrate that the proposed network can classify degraded images over various levels of degradation well. This paper also reveals how the image-quality of training data for a classification network affects the classification performance of degraded images.

26.Anomalous Motion Detection on Highway Using Deep Learning ⬇️

Research in visual anomaly detection draws much interest due to its applications in surveillance. Common datasets for evaluation are constructed using a stationary camera overlooking a region of interest. Previous research has shown promising results in detecting spatial as well as temporal anomalies in these settings. The advent of self-driving cars provides an opportunity to apply visual anomaly detection in a more dynamic application yet no dataset exists in this type of environment. This paper presents a new anomaly detection dataset - the Highway Traffic Anomaly (HTA) dataset - for the problem of detecting anomalous traffic patterns from dash cam videos of vehicles on highways. We evaluate state-of-the-art deep learning anomaly detection models and propose novel variations to these methods. Our results show that state-of-the-art models built for settings with a stationary camera do not translate well to a more dynamic environment. The proposed variations to these SoTA methods show promising results on the new HTA dataset.

27.Geo-PIFu: Geometry and Pixel Aligned Implicit Functions for Single-view Human Reconstruction ⬇️

We propose Geo-PIFu, a method to recover a 3D mesh from a monocular color image of a clothed person. Our method is based on a deep implicit function-based representation to learn latent voxel features using a structure-aware 3D U-Net, to constrain the model in two ways: first, to resolve feature ambiguities in query point encoding, second, to serve as a coarse human shape proxy to regularize the high-resolution mesh and encourage global shape regularity. We show that, by both encoding query points and constraining global shape using latent voxel features, the reconstruction we obtain for clothed human meshes exhibits less shape distortion and improved surface details compared to competing methods. We evaluate Geo-PIFu on a recent human mesh public dataset that is $10 \times$ larger than the private commercial dataset used in PIFu and previous derivative work. On average, we exceed the state of the art by $42.7%$ reduction in Chamfer and Point-to-Surface Distances, and $19.4%$ reduction in normal estimation errors.

28.Multiple Video Frame Interpolation via Enhanced Deformable Separable Convolution ⬇️

Generating non-existing frames from a consecutive video sequence has been an interesting and challenging problem in the video processing field. Recent kernel-based interpolation methods predict pixels with a single convolution process that convolves source frames with spatially adaptive local kernels. However, when scene motion is larger than the pre-defined kernel size, these methods are prone to yield less plausible results and they cannot directly generate a frame at an arbitrary temporal position because the learned kernels are tied to the midpoint in time between the input frames. In this paper, we try to solve these problems and propose a novel approach that we refer to as enhanced deformable separable convolution (EDSC) to estimate not only adaptive kernels, but also offsets, masks and biases to make the network obtain information from non-local neighborhood. During the learning process, different intermediate time step can be involved as a control variable by means of the coord-conv trick, allowing the estimated components to vary with different input temporal information. This makes our method capable to produce multiple in-between frames. Furthermore, we investigate the relationships between our method and other typical kernel- and flow-based methods. Experimental results show that our method performs favorably against the state-of-the-art methods across a broad range of datasets. Code will be publicly available on URL: \url{this https URL}.

29.RasterNet: Modeling Free-Flow Speed using LiDAR and Overhead Imagery ⬇️

Roadway free-flow speed captures the typical vehicle speed in low traffic conditions. Modeling free-flow speed is an important problem in transportation engineering with applications to a variety of design, operation, planning, and policy decisions of highway systems. Unfortunately, collecting large-scale historical traffic speed data is expensive and time consuming. Traditional approaches for estimating free-flow speed use geometric properties of the underlying road segment, such as grade, curvature, lane width, lateral clearance and access point density, but for many roads such features are unavailable. We propose a fully automated approach, RasterNet, for estimating free-flow speed without the need for explicit geometric features. RasterNet is a neural network that fuses large-scale overhead imagery and aerial LiDAR point clouds using a geospatially consistent raster structure. To support training and evaluation, we introduce a novel dataset combining free-flow speeds of road segments, overhead imagery, and LiDAR point clouds across the state of Kentucky. Our method achieves state-of-the-art results on a benchmark dataset.

30.BatVision with GCC-PHAT Features for Better Sound to Vision Predictions ⬇️

Inspired by sophisticated echolocation abilities found in nature, we train a generative adversarial network to predict plausible depth maps and grayscale layouts from sound. To achieve this, our sound-to-vision model processes binaural echo-returns from chirping sounds. We build upon previous work with BatVision that consists of a sound-to-vision model and a self-collected dataset using our mobile robot and low-cost hardware. We improve on the previous model by introducing several changes to the model, which leads to a better depth and grayscale estimation, and increased perceptual quality. Rather than using raw binaural waveforms as input, we generate generalized cross-correlation (GCC) features and use these as input instead. In addition, we change the model generator and base it on residual learning and use spectral normalization in the discriminator. We compare and present both quantitative and qualitative improvements over our previous BatVision model.

31.Road Mapping in Low Data Environments with OpenStreetMap ⬇️

Roads are among the most essential components of any country's infrastructure. By facilitating the movement and exchange of people, ideas, and goods, they support economic and cultural activity both within and across local and international borders. A comprehensive, up-to-date mapping of the geographical distribution of roads and their quality thus has the potential to act as an indicator for broader economic development. Such an indicator has a variety of high-impact applications, particularly in the planning of rural development projects where up-to-date infrastructure information is not available. This work investigates the viability of high resolution satellite imagery and crowd-sourced resources like OpenStreetMap in the construction of such a mapping. We experiment with state-of-the-art deep learning methods to explore the utility of OpenStreetMap data in road classification and segmentation tasks. We also compare the performance of models in different mask occlusion scenarios as well as out-of-country domains. Our comparison raises important pitfalls to consider in image-based infrastructure classification tasks, and shows the need for local training data specific to regions of interest for reliable performance.

32.Emergent Properties of Foveated Perceptual Systems ⬇️

We introduce foveated perceptual systems, inspired by human biological systems, and examine the impact that this foveation stage has on the nature and robustness of subsequently learned visual representation. Specifically, these \textit{two-stage} perceptual systems first foveate an image, inducing a texture-like encoding of peripheral information, which is then inputted to a convolutional neural network (CNN) and trained to perform scene categorization. We find that: 1-- Systems trained on foveated inputs (Foveation-Nets) have similar generalization as networks trained on matched-resource networks without foveated input (Standard-Nets), yet show greater cross-generalization. 2-- Foveation-Nets show higher robustness than Standard-Nets to scotoma (fovea removed) occlusions, driven by the first foveation stage. 3-- Subsequent representations learned in the CNN of Foveation-Nets weigh center information more strongly than Standard-Nets. 4-- Foveation-Nets show less sensitivity to low-spatial frequency information than Standard-Nets. Furthermore, when we added biological and artificial augmentation mechanisms to each system through simulated eye-movements or random cropping and mirroring respectively, we found that these effects were amplified. Taken together, we find evidence that foveated perceptual systems learn a visual representation that is distinct from non-foveated perceptual systems, with implications in generalization, robustness, and perceptual sensitivity. These results provide computational support for the idea that the foveated nature of the human visual system might confer a functional advantage for scene representation.

33.GradAug: A New Regularization Method for Deep Neural Networks ⬇️

We propose a new regularization method to alleviate over-fitting in deep neural networks. The key idea is utilizing randomly transformed training samples to regularize a set of sub-networks, which are originated by sampling the width of the original network, in the training process. As such, the proposed method introduces self-guided disturbances to the raw gradients of the network and therefore is termed as Gradient Augmentation (GradAug). We demonstrate that GradAug can help the network learn well-generalized and more diverse representations. Moreover, it is easy to implement and can be applied to various structures and applications. GradAug improves ResNet-50 to 78.79% on ImageNet classification, which is a new state-of-the-art accuracy. By combining with CutMix, it further boosts the performance to 79.58%, which outperforms an ensemble of advanced training tricks. The generalization ability is evaluated on COCO object detection and instance segmentation where GradAug significantly surpasses other state-of-the-art methods. GradAug is also robust to image distortions and adversarial attacks and is highly effective in the low data regimes.

34.ShapeFlow: Learnable Deformations Among 3D Shapes ⬇️

We present ShapeFlow, a flow-based model for learning a deformation space for entire classes of 3D shapes with large intra-class variations. ShapeFlow allows learning a multi-template deformation space that is agnostic to shape topology, yet preserves fine geometric details. Different from a generative space where a latent vector is directly decoded into a shape, a deformation space decodes a vector into a continuous flow that can advect a source shape towards a target. Such a space naturally allows the disentanglement of geometric style (coming from the source) and structural pose (conforming to the target). We parametrize the deformation between geometries as a learned continuous flow field via a neural network and show that such deformations can be guaranteed to have desirable properties, such as be bijectivity, freedom from self-intersections, or volume preservation. We illustrate the effectiveness of this learned deformation space for various downstream applications, including shape generation via deformation, geometric style transfer, unsupervised learning of a consistent parameterization for entire classes of shapes, and shape interpolation.

35.Geodesic-HOF: 3D Reconstruction Without Cutting Corners ⬇️

Single-view 3D object reconstruction is a challenging fundamental problem in computer vision, largely due to the morphological diversity of objects in the natural world. In particular, high curvature regions are not always captured effectively by methods trained using only set-based loss functions, resulting in reconstructions short-circuiting the surface or cutting corners. In particular, high curvature regions are not always captured effectively by methods trained using only set-based loss functions, resulting in reconstructions short-circuiting the surface or cutting corners. To address this issue, we propose learning an image-conditioned mapping function from a canonical sampling domain to a high dimensional space where the Euclidean distance is equal to the geodesic distance on the object. The first three dimensions of a mapped sample correspond to its 3D coordinates. The additional lifted components contain information about the underlying geodesic structure. Our results show that taking advantage of these learned lifted coordinates yields better performance for estimating surface normals and generating surfaces than using point cloud reconstructions alone. Further, we find that this learned geodesic embedding space provides useful information for applications such as unsupervised object decomposition.

36.Actor-Context-Actor Relation Network for Spatio-Temporal Action Localization ⬇️

Localizing persons and recognizing their actions from videos is a challenging task towards high-level video understanding. Recent advances have been achieved by modeling either 'actor-actor' or 'actor-context' relations. However, such direct first-order relations are not sufficient for localizing actions in complicated scenes. Some actors might be indirectly related via objects or background context in the scene. Such indirect relations are crucial for determining the action labels but are mostly ignored by existing work. In this paper, we propose to explicitly model the Actor-Context-Actor Relation, which can capture indirect high-order supportive information for effectively reasoning actors' actions in complex scenes. To this end, we design an Actor-Context-Actor Relation Network (ACAR-Net) which builds upon a novel High-order Relation Reasoning Operator to model indirect relations for spatio-temporal action localization. Moreover, to allow utilizing more temporal contexts, we extend our framework with an Actor-Context Feature Bank for reasoning long-range high-order relations. Extensive experiments on AVA dataset validate the effectiveness of our ACAR-Net. Ablation studies show the advantages of modeling high-order relations over existing first-order relation reasoning methods. The proposed ACAR-Net is also the core module of our 1st place solution in AVA-Kinetics Crossover Challenge 2020. Training code and models will be available at this https URL.

37.Meta Approach to Data Augmentation Optimization ⬇️

Data augmentation policies drastically improve the performance of image recognition tasks, especially when the policies are optimized for the target data and tasks. In this paper, we propose to optimize image recognition models and data augmentation policies simultaneously to improve the performance using gradient descent. Unlike prior methods, our approach avoids using proxy tasks or reducing search space, and can directly improve the validation performance. Our method achieves efficient and scalable training by approximating the gradient of policies by implicit gradient with Neumann series approximation. We demonstrate that our approach can improve the performance of various image classification tasks, including ImageNet classification and fine-grained recognition, without using dataset-specific hyperparameter tuning.

38.Team RUC_AIM3 Technical Report at Activitynet 2020 Task 2: Exploring Sequential Events Detection for Dense Video Captioning ⬇️

Detecting meaningful events in an untrimmed video is essential for dense video captioning. In this work, we propose a novel and simple model for event sequence generation and explore temporal relationships of the event sequence in the video. The proposed model omits inefficient two-stage proposal generation and directly generates event boundaries conditioned on bi-directional temporal dependency in one pass. Experimental results show that the proposed event sequence generation model can generate more accurate and diverse events within a small number of proposals. For the event captioning, we follow our previous work to employ the intra-event captioning models into our pipeline system. The overall system achieves state-of-the-art performance on the dense-captioning events in video task with 9.894 METEOR score on the challenge testing set.

39.Optical Music Recognition: State of the Art and Major Challenges ⬇️

Optical Music Recognition (OMR) is concerned with transcribing sheet music into a machine-readable format. The transcribed copy should allow musicians to compose, play and edit music by taking a picture of a music sheet. Complete transcription of sheet music would also enable more efficient archival. OMR facilitates examining sheet music statistically or searching for patterns of notations, thus helping use cases in digital musicology too. Recently, there has been a shift in OMR from using conventional computer vision techniques towards a deep learning approach. In this paper, we review relevant works in OMR, including fundamental methods and significant outcomes, and highlight different stages of the OMR pipeline. These stages often lack standard input and output representation and standardised evaluation. Therefore, comparing different approaches and evaluating the impact of different processing methods can become rather complex. This paper provides recommendations for future work, addressing some of the highlighted issues and represents a position in furthering this important field of research.

40.FenceMask: A Data Augmentation Approach for Pre-extracted Image Features ⬇️

We propose a novel data augmentation method named 'FenceMask' that exhibits outstanding performance in various computer vision tasks. It is based on the 'simulation of object occlusion' strategy, which aim to achieve the balance between object occlusion and information retention of the input data. By enhancing the sparsity and regularity of the occlusion block, our augmentation method overcome the difficulty of small object augmentation and notably improve performance over baselines. Sufficient experiments prove the performance of our method is better than other simulate object occlusion approaches. We tested it on CIFAR10, CIFAR100 and ImageNet datasets for Coarse-grained classification, COCO2017 and VisDrone datasets for detection, Oxford Flowers, Cornel Leaf and Stanford Dogs datasets for Fine-Grained Visual Categorization. Our method achieved significant performance improvement on Fine-Grained Visual Categorization task and VisDrone dataset.

41.Explicitly Modeled Attention Maps for Image Classification ⬇️

Self-attention networks have shown remarkable progress in computer vision tasks such as image classification. The main benefit of the self-attention mechanism is the ability to capture long-range feature interactions in attention-maps. However, the computation of attention-maps requires a learnable key, query, and positional encoding, whose usage is often not intuitive and computationally expensive. To mitigate this problem, we propose a novel self-attention module with explicitly modeled attention-maps using only a single learnable parameter for low computational overhead. The design of explicitly modeled attention-maps using geometric prior is based on the observation that the spatial context for a given pixel within an image is mostly dominated by its neighbors, while more distant pixels have a minor contribution. Concretely, the attention-maps are parametrized via simple functions (e.g., Gaussian kernel) with a learnable radius, which is modeled independently of the input content. Our evaluation shows that our method achieves an accuracy improvement of up to 2.2% over the ResNet-baselines in ImageNet ILSVRC and outperforms other self-attention methods such as AA-ResNet152 (Bello et al., 2019) in accuracy by 0.9% with 6.4% fewer parameters and 6.7% fewer GFLOPs.

42.Cityscapes 3D: Dataset and Benchmark for 9 DoF Vehicle Detection ⬇️

Detecting vehicles and representing their position and orientation in the three dimensional space is a key technology for autonomous driving. Recently, methods for 3D vehicle detection solely based on monocular RGB images gained popularity. In order to facilitate this task as well as to compare and drive state-of-the-art methods, several new datasets and benchmarks have been published. Ground truth annotations of vehicles are usually obtained using lidar point clouds, which often induces errors due to imperfect calibration or synchronization between both sensors. To this end, we propose Cityscapes 3D, extending the original Cityscapes dataset with 3D bounding box annotations for all types of vehicles. In contrast to existing datasets, our 3D annotations were labeled using stereo RGB images only and capture all nine degrees of freedom. This leads to a pixel-accurate reprojection in the RGB image and a higher range of annotations compared to lidar-based approaches. In order to ease multitask learning, we provide a pairing of 2D instance segments with 3D bounding boxes. In addition, we complement the Cityscapes benchmark suite with 3D vehicle detection based on the new annotations as well as metrics presented in this work. Dataset and benchmark are available online.

43.An adversarial learning algorithm for mitigating gender bias in face recognition ⬇️

State-of-the-art face recognition networks implicitly encode gender information while being trained for identity classification. Gender is often viewed as an important face attribute to recognize humans. But, the expression of gender information in deep facial features appears to contribute to gender bias in face recognition, i.e. we find a significant difference in the recognition accuracy of DCNNs on male and female faces. We hypothesize that reducing implicitly encoded gender information will help reduce this gender bias. Therefore, we present a novel approach called `Adversarial Gender De-biasing (AGD)' to reduce the strength of gender information in face recognition features. We accomplish this by introducing a bias reducing classification loss $L_{br}$. We show that AGD significantly reduces bias, while achieving reasonable recognition performance. The results of our approach are presented on two state-of-the-art networks.

44.A Generalized Asymmetric Dual-front Model for Active Contours and Image Segmentation ⬇️

The geodesic distance-based dual-front curve evolution model is a powerful and efficient solution to the active contours and image segmentation issues. In its basic formulation, the dual-front model regards the meeting interfaces of two adjacent Voronoi regions as the evolving curves in the course of curve evolution. One of the most crucial ingredients for the construction of Voronoi regions or Voronoi diagram is the geodesic metrics and the corresponding geodesic distance. In this paper, we introduce a new type of geodesic metrics that encodes the edge-based anisotropy features, the region-based homogeneity penalization and asymmetric enhancement. In contrast to the original isotropic dual-front model, the use of the asymmetric enhancement can reduce the risk of shortcuts or leakage problems especially when the initial curves are far away from the target boundaries. Moreover, the proposed dual-front model can be applied for image segmentation in conjunction with various region-based homogeneity terms, whereas the original model only makes use of the piecewise constant case. The numerical experiments on both synthetic and real images show that the proposed model indeed achieves encouraging results.

45.Multi-Miner: Object-Adaptive Region Mining for Weakly-Supervised Semantic Segmentation ⬇️

Object region mining is a critical step for weakly-supervised semantic segmentation. Most recent methods mine the object regions by expanding the seed regions localized by class activation maps. They generally do not consider the sizes of objects and apply a monotonous procedure to mining all the object regions. Thus their mined regions are often insufficient in number and scale for large objects, and on the other hand easily contaminated by surrounding backgrounds for small objects. In this paper, we propose a novel multi-miner framework to perform a region mining process that adapts to diverse object sizes and is thus able to mine more integral and finer object regions. Specifically, our multi-miner leverages a parallel modulator to check whether there are remaining object regions for each single object, and guide a category-aware generator to mine the regions of each object independently. In this way, the multi-miner adaptively takes more steps for large objects and fewer steps for small objects. Experiment results demonstrate that the multi-miner offers better region mining results and helps achieve better segmentation performance than state-of-the-art weakly-supervised semantic segmentation methods.

46.On Saliency Maps and Adversarial Robustness ⬇️

A Very recent trend has emerged to couple the notion of interpretability and adversarial robustness, unlike earlier efforts which solely focused on good interpretations or robustness against adversaries. Works have shown that adversarially trained models exhibit more interpretable saliency maps than their non-robust counterparts, and that this behavior can be quantified by considering the alignment between input image and saliency map. In this work, we provide a different perspective to this coupling, and provide a method, Saliency based Adversarial training (SAT), to use saliency maps to improve adversarial robustness of a model. In particular, we show that using annotations such as bounding boxes and segmentation masks, already provided with a dataset, as weak saliency maps, suffices to improve adversarial robustness with no additional effort to generate the perturbations themselves. Our empirical results on CIFAR-10, CIFAR-100, Tiny ImageNet and Flower-17 datasets consistently corroborate our claim, by showing improved adversarial robustness using our method. saliency maps. We also show how using finer and stronger saliency maps leads to more robust models, and how integrating SAT with existing adversarial training methods, further boosts performance of these existing methods.

47.PCAAE: Principal Component Analysis Autoencoder for organising the latent space of generative networks ⬇️

Autoencoders and generative models produce some of the most spectacular deep learning results to date. However, understanding and controlling the latent space of these models presents a considerable challenge. Drawing inspiration from principal component analysis and autoencoder, we propose the Principal Component Analysis Autoencoder (PCAAE). This is a novel autoencoder whose latent space verifies two properties. Firstly, the dimensions are organised in decreasing importance with respect to the data at hand. Secondly, the components of the latent space are statistically independent. We achieve this by progressively increasing the latent space during training, and with a covariance loss applied to the latent codes. The resulting autoencoder produces a latent space which separates the intrinsic attributes of the data into different components of the latent space, in a completely unsupervised manner. We also describe an extension of our approach to the case of powerful, pre-trained GANs. We show results on both synthetic examples of shapes and on a state-of-the-art GAN. For example, we are able to separate the color shade scale of hair and skin, pose of faces and the gender in the CelebA, without accessing any labels. We compare the PCAAE with other state-of-the-art approaches, in particular with respect to the ability to disentangle attributes in the latent space. We hope that this approach will contribute to better understanding of the intrinsic latent spaces of powerful deep generative models.

48.Few-shot Object Detection on Remote Sensing Images ⬇️

In this paper, we deal with the problem of object detection on remote sensing images. Previous methods have developed numerous deep CNN-based methods for object detection on remote sensing images and the report remarkable achievements in detection performance and efficiency. However, current CNN-based methods mostly require a large number of annotated samples to train deep neural networks and tend to have limited generalization abilities for unseen object categories. In this paper, we introduce a few-shot learning-based method for object detection on remote sensing images where only a few annotated samples are provided for the unseen object categories. More specifically, our model contains three main components: a meta feature extractor that learns to extract feature representations from input images, a reweighting module that learn to adaptively assign different weights for each feature representation from the support images, and a bounding box prediction module that carries out object detection on the reweighted feature maps. We build our few-shot object detection model upon YOLOv3 architecture and develop a multi-scale object detection framework. Experiments on two benchmark datasets demonstrate that with only a few annotated samples our model can still achieve a satisfying detection performance on remote sensing images and the performance of our model is significantly better than the well-established baseline models.

49.Working with scale: 2nd place solution to Product Detection in Densely Packed Scenes [Technical Report] ⬇️

This report describes a 2nd place solution of the detection challenge which is held within CVPR 2020 Retail-Vision workshop. Instead of going further considering previous results this work mainly aims to verify previously observed takeaways by re-experimenting. The reliability and reproducibility of the results are reached by incorporating a popular object detection toolbox - MMDetection. In this report, I firstly represent the results received for Faster-RCNN and RetinaNet models, which were taken for comparison in the original work. Then I describe the experiment results with more advanced models. The final section reviews two simple tricks for Faster-RCNN model that were used for my final submission: changing default anchor scale parameter and train-time image tiling. The source code is available at this https URL.

50.Adaptively Meshed Video Stabilization ⬇️

Video stabilization is essential for improving visual quality of shaky videos. The current video stabilization methods usually take feature trajectories in the background to estimate one global transformation matrix or several transformation matrices based on a fixed mesh, and warp shaky frames into their stabilized views. However, these methods may not model the shaky camera motion well in complicated scenes, such as scenes containing large foreground objects or strong parallax, and may result in notable visual artifacts in the stabilized videos. To resolve the above issues, this paper proposes an adaptively meshed method to stabilize a shaky video based on all of its feature trajectories and an adaptive blocking strategy. More specifically, we first extract feature trajectories of the shaky video and then generate a triangle mesh according to the distribution of the feature trajectories in each frame. Then transformations between shaky frames and their stabilized views over all triangular grids of the mesh are calculated to stabilize the shaky video. Since more feature trajectories can usually be extracted from all regions, including both background and foreground regions, a finer mesh will be obtained and provided for camera motion estimation and frame warping. We estimate the mesh-based transformations of each frame by solving a two-stage optimization problem. Moreover, foreground and background feature trajectories are no longer distinguished and both contribute to the estimation of the camera motion in the proposed optimization problem, which yields better estimation performance than previous works, particularly in challenging videos with large foreground objects or strong parallax.

51.Alternating ConvLSTM: Learning Force Propagation with Alternate State Updates ⬇️

Data-driven simulation is an important step-forward in computational physics when traditional numerical methods meet their limits. Learning-based simulators have been widely studied in past years; however, most previous works view simulation as a general spatial-temporal prediction problem and take little physical guidance in designing their neural network architectures. In this paper, we introduce the alternating convolutional Long Short-Term Memory (Alt-ConvLSTM) that models the force propagation mechanisms in a deformable object with near-uniform material properties. Specifically, we propose an accumulation state, and let the network update its cell state and the accumulation state alternately. We demonstrate how this novel scheme imitates the alternate updates of the first and second-order terms in the forward Euler method of numerical PDE solvers. Benefiting from this, our network only requires a small number of parameters, independent of the number of the simulated particles, and also retains the essential features in ConvLSTM, making it naturally applicable to sequential data with spatial inputs and outputs. We validate our Alt-ConvLSTM on human soft tissue simulation with thousands of particles and consistent body pose changes. Experimental results show that Alt-ConvLSTM efficiently models the material kinetic features and greatly outperforms vanilla ConvLSTM with only the single state update.

52.2D Image Relighting with Image-to-Image Translation ⬇️

With the advent of Generative Adversarial Networks (GANs), a finer level of control in manipulating various features of an image has become possible. One example of such fine manipulation is changing the position of the light source in a scene. This is fundamentally an ill-posed problem, since it requires understanding the scene geometry to generate proper lighting effects. This problem is not a trivial one and can become even more complicated if we want to change the direction of the light source from any direction to a specific one. Here we provide our attempt to solve this problem using GANs. Specifically, pix2pix [arXiv:1611.07004] trained with the dataset VIDIT [arXiv:2005.05460] which contains images of the same scene with different types of light temperature and 8 different light source positions (N, NE, E, SE, S, SW, W, NW). The results are 8 neural networks trained to be able to change the direction of the light source from any direction to one of the 8 previously mentioned. Additionally, we provide, as a tool, a simple CNN trained to identify the direction of the light source in an image.

53.Disentanglement for Discriminative Visual Recognition ⬇️

Recent successes of deep learning-based recognition rely on maintaining the content related to the main-task label. However, how to explicitly dispel the noisy signals for better generalization in a controllable manner remains an open issue. For instance, various factors such as identity-specific attributes, pose, illumination and expression affect the appearance of face images. Disentangling the identity-specific factors is potentially beneficial for facial expression recognition (FER). This chapter systematically summarize the detrimental factors as task-relevant/irrelevant semantic variations and unspecified latent variation. In this chapter, these problems are casted as either a deep metric learning problem or an adversarial minimax game in the latent space. For the former choice, a generalized adaptive (N+M)-tuplet clusters loss function together with the identity-aware hard-negative mining and online positive mining scheme can be used for identity-invariant FER. The better FER performance can be achieved by combining the deep metric loss and softmax loss in a unified two fully connected layer branches framework via joint optimization. For the latter solution, it is possible to equipping an end-to-end conditional adversarial network with the ability to decompose an input sample into three complementary parts. The discriminative representation inherits the desired invariance property guided by prior knowledge of the task, which is marginal independent to the task-relevant/irrelevant semantic and latent variations. The framework achieves top performance on a serial of tasks, including lighting, makeup, disguise-tolerant face recognition and facial attributes recognition. This chapter systematically summarize the popular and practical solution for disentanglement to achieve more discriminative visual recognition.

54.ReLGAN: Generalization of Consistency for GAN with Disjoint Constraints and Relative Learning of Generative Processes for Multiple Transformation Learning ⬇️

Image to image transformation has gained popularity from different research communities due to its enormous impact on different applications, including medical. In this work, we have introduced a generalized scheme for consistency for GAN architectures with two new concepts of Transformation Learning (TL) and Relative Learning (ReL) for enhanced learning image transformations. Consistency for GAN architectures suffered from inadequate constraints and failed to learn multiple and multi-modal transformations, which is inevitable for many medical applications. The main drawback is that it focused on creating an intermediate and workable hybrid, which is not permissible for the medical applications which focus on minute details. Another drawback is the weak interrelation between the two learning phases and TL and ReL have introduced improved coordination among them. We have demonstrated the capability of the novel network framework on public datasets. We emphasized that our novel architecture produced an improved neural image transformation version for the image, which is more acceptable to the medical community. Experiments and results demonstrated the effectiveness of our framework with enhancement compared to the previous works.

55.Relative Pose Estimation for Stereo Rolling Shutter Cameras ⬇️

In this paper, we present a novel linear algorithm to estimate the 6 DoF relative pose from consecutive frames of stereo rolling shutter (RS) cameras. Our method is derived based on the assumption that stereo cameras undergo motion with constant velocity around the center of the baseline, which needs 9 pairs of correspondences on both left and right consecutive frames. The stereo RS images enable the recovery of depth maps from the semi-global matching (SGM) algorithm. With the estimated camera motion and depth map, we can correct the RS images to get the undistorted images without any scene structure assumption. Experiments on both simulated points and synthetic RS images demonstrate the effectiveness of our algorithm in relative pose estimation.

56.Geometry-Aware Instance Segmentation with Disparity Maps ⬇️

Most previous works of outdoor instance segmentation for images only use color information. We explore a novel direction of sensor fusion to exploit stereo cameras. Geometric information from disparities helps separate overlapping objects of the same or different classes. Moreover, geometric information penalizes region proposals with unlikely 3D shapes thus suppressing false positive detections. Mask regression is based on 2D, 2.5D, and 3D ROI using the pseudo-lidar and image-based representations. These mask predictions are fused by a mask scoring process. However, public datasets only adopt stereo systems with shorter baseline and focal legnth, which limit measuring ranges of stereo cameras. We collect and utilize High-Quality Driving Stereo (HQDS) dataset, using much longer baseline and focal length with higher resolution. Our performance attains state of the art. Please refer to our project page. The full paper is available here.

57.Hyper RPCA: Joint Maximum Correntropy Criterion and Laplacian Scale Mixture Modeling On-the-Fly for Moving Object Detection ⬇️

Moving object detection is critical for automated video analysis in many vision-related tasks, such as surveillance tracking, video compression coding, etc. Robust Principal Component Analysis (RPCA), as one of the most popular moving object modelling methods, aims to separate the temporally varying (i.e., moving) foreground objects from the static background in video, assuming the background frames to be low-rank while the foreground to be spatially sparse. Classic RPCA imposes sparsity of the foreground component using l1-norm, and minimizes the modeling error via 2-norm. We show that such assumptions can be too restrictive in practice, which limits the effectiveness of the classic RPCA, especially when processing videos with dynamic background, camera jitter, camouflaged moving object, etc. In this paper, we propose a novel RPCA-based model, called Hyper RPCA, to detect moving objects on the fly. Different from classic RPCA, the proposed Hyper RPCA jointly applies the maximum correntropy criterion (MCC) for the modeling error, and Laplacian scale mixture (LSM) model for foreground objects. Extensive experiments have been conducted, and the results demonstrate that the proposed Hyper RPCA has competitive performance for foreground detection to the state-of-the-art algorithms on several well-known benchmark datasets.

58.Generative 3D Part Assembly via Dynamic Graph Learning ⬇️

Autonomous part assembly is a challenging yet crucial task in 3D computer vision and robotics. Analogous to buying an IKEA furniture, given a set of 3D parts that can assemble a single shape, an intelligent agent needs to perceive the 3D part geometry, reason to propose pose estimations for the input parts, and finally call robotic planning and control routines for actuation. In this paper, we focus on the pose estimation subproblem from the vision side involving geometric and relational reasoning over the input part geometry. Essentially, the task of generative 3D part assembly is to predict a 6-DoF part pose, including a rigid rotation and translation, for each input part that assembles a single 3D shape as the final output. To tackle this problem, we propose an assembly-oriented dynamic graph learning framework that leverages an iterative graph neural network as a backbone. It explicitly conducts sequential part assembly refinements in a coarse-to-fine manner, exploits a pair of part relation reasoning module and part aggregation module for dynamically adjusting both part features and their relations in the part graph. We conduct extensive experiments and quantitative comparisons to three strong baseline methods, demonstrating the effectiveness of the proposed approach.

59.PrimA6D: Rotational Primitive Reconstruction for Enhanced and Robust 6D Pose Estimation ⬇️

In this paper, we introduce a rotational primitive prediction based 6D object pose estimation using a single image as an input. We solve for the 6D object pose of a known object relative to the camera using a single image with occlusion. Many recent state-of-the-art (SOTA) two-step approaches have exploited image keypoints extraction followed by PnP regression for pose estimation. Instead of relying on bounding box or keypoints on the object, we propose to learn orientation-induced primitive so as to achieve the pose estimation accuracy regardless of the object size. We leverage a Variational AutoEncoder (VAE) to learn this underlying primitive and its associated keypoints. The keypoints inferred from the reconstructed primitive image are then used to regress the rotation using PnP. Lastly, we compute the translation in a separate localization module to complete the entire 6D pose estimation. When evaluated over public datasets, the proposed method yields a notable improvement over the LINEMOD, the Occlusion LINEMOD, and the YCB-Video dataset. We further provide a synthetic-only trained case presenting comparable performance to the existing methods which require real images in the training phase.

60.Cascaded deep monocular 3D human pose estimation with evolutionary training data ⬇️

End-to-end deep representation learning has achieved remarkable accuracy for monocular 3D human pose estimation, yet these models may fail for unseen poses with limited and fixed training data. This paper proposes a novel data augmentation method that: (1) is scalable for synthesizing massive amount of training data (over 8 million valid 3D human poses with corresponding 2D projections) for training 2D-to-3D networks, (2) can effectively reduce dataset bias. Our method evolves a limited dataset to synthesize unseen 3D human skeletons based on a hierarchical human representation and heuristics inspired by prior knowledge. Extensive experiments show that our approach not only achieves state-of-the-art accuracy on the largest public benchmark, but also generalizes significantly better to unseen and rare poses. Relevant files and tools are available at the project website.

61.Domain Adaptation and Image Classification via Deep Conditional Adaptation Network ⬇️

Unsupervised domain adaptation aims to generalize the supervised model trained on a source domain to an unlabeled target domain. Marginal distribution alignment of feature spaces is widely used to reduce the domain discrepancy between the source and target domains. However, it assumes that the source and target domains share the same label distribution, which limits their application scope. In this paper, we consider a more general application scenario where the label distributions of the source and target domains are not the same. In this scenario, marginal distribution alignment-based methods will be vulnerable to negative transfer. To address this issue, we propose a novel unsupervised domain adaptation method, Deep Conditional Adaptation Network (DCAN), based on conditional distribution alignment of feature spaces. To be specific, we reduce the domain discrepancy by minimizing the Conditional Maximum Mean Discrepancy between the conditional distributions of deep features on the source and target domains, and extract the discriminant information from target domain by maximizing the mutual information between samples and the prediction labels. In addition, DCAN can be used to address a special scenario, Partial unsupervised domain adaptation, where the target domain category is a subset of the source domain category. Experiments on both unsupervised domain adaptation and Partial unsupervised domain adaptation show that DCAN achieves superior classification performance over state-of-the-art methods. In particular, DCAN achieves great improvement in the tasks with large difference in label distributions (6.1% on SVHN to MNIST, 5.4% in UDA tasks on Office-Home and 4.5% in Partial UDA tasks on Office-Home).

62.Recurrent Distillation based Crowd Counting ⬇️

In recent years, with the progress of deep learning technologies, crowd counting has been rapidly developed. In this work, we propose a simple yet effective crowd counting framework that is able to achieve the state-of-the-art performance on various crowded scenes. In particular, we first introduce a perspective-aware density map generation method that is able to produce ground-truth density maps from point annotations to train crowd counting model to accomplish superior performance than prior density map generation techniques. Besides, leveraging our density map generation method, we propose an iterative distillation algorithm to progressively enhance our model with identical network structures, without significantly sacrificing the dimension of the output density maps. In experiments, we demonstrate that, with our simple convolutional neural network architecture strengthened by our proposed training algorithm, our model is able to outperform or be comparable with the state-of-the-art methods. Furthermore, we also evaluate our density map generation approach and distillation algorithm in ablation studies.

63.3D Reconstruction of Novel Object Shapes from Single Images ⬇️

The key challenge in single image 3D shape reconstruction is to ensure that deep models can generalize to shapes which were not part of the training set. This is difficult because the algorithm must infer the occluded portion of the surface by leveraging the shape characteristics of the training data, and can therefore be vulnerable to overfitting. Such generalization to unseen categories of objects is a function of architecture design and training approaches. This paper introduces SDFNet, a novel shape prediction architecture and training approach which supports effective generalization. We provide an extensive investigation of the factors which influence generalization accuracy and its measurement, ranging from the consistent use of 3D shape metrics to the choice of rendering approach and the large-scale evaluation on unseen shapes using ShapeNetCore.v2 and ABC. We show that SDFNet provides state-of-the-art performance on seen and unseen shapes relative to existing baseline methods GenRe and OccNet. We provide the first large-scale experimental evaluation of generalization performance. The codebase released with this article will allow for the consistent evaluation and comparison of methods for single image shape reconstruction.

64.Exploiting the ConvLSTM: Human Action Recognition using Raw Depth Video-Based Recurrent Neural Networks ⬇️

As in many other different fields, deep learning has become the main approach in most computer vision applications, such as scene understanding, object recognition, computer-human interaction or human action recognition (HAR). Research efforts within HAR have mainly focused on how to efficiently extract and process both spatial and temporal dependencies of video sequences. In this paper, we propose and compare, two neural networks based on the convolutional long short-term memory unit, namely ConvLSTM, with differences in the architecture and the long-term learning strategy. The former uses a video-length adaptive input data generator (\emph{stateless}) whereas the latter explores the \emph{stateful} ability of general recurrent neural networks but applied in the particular case of HAR. This stateful property allows the model to accumulate discriminative patterns from previous frames without compromising computer memory. Experimental results on the large-scale NTU RGB+D dataset show that the proposed models achieve competitive recognition accuracies with lower computational cost compared with state-of-the-art methods and prove that, in the particular case of videos, the rarely-used stateful mode of recurrent neural networks significantly improves the accuracy obtained with the standard mode. The recognition accuracies obtained are 75.26% (CS) and 75.45% (CV) for the stateless model, with an average time consumption per video of 0.21 s, and 80.43% (CS) and 79.91%(CV) with 0.89 s for the stateful version.

65.3DFCNN: Real-Time Action Recognition using 3D Deep Neural Networks with Raw Depth Information ⬇️

Human actions recognition is a fundamental task in artificial vision, that has earned a great importance in recent years due to its multiple applications in different areas. %, such as the study of human behavior, security or video surveillance. In this context, this paper describes an approach for real-time human action recognition from raw depth image-sequences, provided by an RGB-D camera. The proposal is based on a 3D fully convolutional neural network, named 3DFCNN, which automatically encodes spatio-temporal patterns from depth sequences without %any costly pre-processing. Furthermore, the described 3D-CNN allows %automatic features extraction and actions classification from the spatial and temporal encoded information of depth sequences. The use of depth data ensures that action recognition is carried out protecting people's privacy% allows recognizing the actions carried out by people, protecting their privacy%\sout{of them} , since their identities can not be recognized from these data. %\st{ from depth images.} 3DFCNN has been evaluated and its results compared to those from other state-of-the-art methods within three widely used %large-scale NTU RGB+D datasets, with different characteristics (resolution, sensor type, number of views, camera location, etc.). The obtained results allows validating the proposal, concluding that it outperforms several state-of-the-art approaches based on classical computer vision techniques. Furthermore, it achieves action recognition accuracy comparable to deep learning based state-of-the-art methods with a lower computational cost, which allows its use in real-time applications.

66.Split-Merge Pooling ⬇️

There are a variety of approaches to obtain a vast receptive field with convolutional neural networks (CNNs), such as pooling or striding convolutions. Most of these approaches were initially designed for image classification and later adapted to dense prediction tasks, such as semantic segmentation. However, the major drawback of this adaptation is the loss of spatial information. Even the popular dilated convolution approach, which in theory is able to operate with full spatial resolution, needs to subsample features for large image sizes in order to make the training and inference tractable. In this work, we introduce Split-Merge pooling to fully preserve the spatial information without any subsampling. By applying Split-Merge pooling to deep networks, we achieve, at the same time, a very large receptive field. We evaluate our approach for dense semantic segmentation of large image sizes taken from the Cityscapes and GTA-5 datasets. We demonstrate that by replacing max-pooling and striding convolutions with our split-merge pooling, we are able to improve the accuracy of different variations of ResNet significantly.

67.V2E: From video frames to realistic DVS event camera streams ⬇️

To help meet the increasing need for dynamic vision sensor (DVS) event camera data, we developed the v2e toolbox, which generates synthetic DVS event streams from intensity frame videos. Videos can be of any type, either real or synthetic. v2e optionally uses synthetic slow motion to upsample the video frame rate and then generates DVS events from these frames using a realistic pixel model that includes event threshold mismatch, finite illumination-dependent bandwidth, and several types of noise. v2e includes an algorithm that determines the DVS thresholds and bandwidth so that the synthetic event stream statistics match a given reference DVS recording. v2e is the first toolbox that can synthesize realistic low light DVS data. This paper also clarifies misleading claims about DVS characteristics in some of the computer vision literature. The v2e website is this https URL and code is hosted at this https URL.

68.Sensorless Freehand 3D Ultrasound Reconstruction via Deep Contextual Learning ⬇️

Transrectal ultrasound (US) is the most commonly used imaging modality to guide prostate biopsy and its 3D volume provides even richer context information. Current methods for 3D volume reconstruction from freehand US scans require external tracking devices to provide spatial position for every frame. In this paper, we propose a deep contextual learning network (DCL-Net), which can efficiently exploit the image feature relationship between US frames and reconstruct 3D US volumes without any tracking device. The proposed DCL-Net utilizes 3D convolutions over a US video segment for feature extraction. An embedded self-attention module makes the network focus on the speckle-rich areas for better spatial movement prediction. We also propose a novel case-wise correlation loss to stabilize the training process for improved accuracy. Highly promising results have been obtained by using the developed method. The experiments with ablation studies demonstrate superior performance of the proposed method by comparing against other state-of-the-art methods. Source code of this work is publicly available at this https URL.

69.Uncertainty-aware Score Distribution Learning for Action Quality Assessment ⬇️

Assessing action quality from videos has attracted growing attention in recent years. Most existing approaches usually tackle this problem based on regression algorithms, which ignore the intrinsic ambiguity in the score labels caused by multiple judges or their subjective appraisals. To address this issue, we propose an uncertainty-aware score distribution learning (USDL) approach for action quality assessment (AQA). Specifically, we regard an action as an instance associated with a score distribution, which describes the probability of different evaluated scores. Moreover, under the circumstance where fine-grained score labels are available (e.g., difficulty degree of an action or multiple scores from different judges), we further devise a multi-path uncertainty-aware score distributions learning (MUSDL) method to explore the disentangled components of a score. We conduct experiments on three AQA datasets containing various Olympic actions and surgical activities, where our approaches set new state-of-the-arts under the Spearman's Rank Correlation.

70.Convolutional Generation of Textured 3D Meshes ⬇️

Recent generative models for 2D images achieve impressive visual results, but clearly lack the ability to perform 3D reasoning. This heavily restricts the degree of control over generated objects as well as the possible applications of such models. In this work, we leverage recent advances in differentiable rendering to design a framework that can generate triangle meshes and associated high-resolution texture maps, using only 2D supervision from single-view natural images. A key contribution of our work is the encoding of the mesh and texture as 2D representations, which are semantically aligned and can be easily modeled by a 2D convolutional GAN. We demonstrate the efficacy of our method on Pascal3D+ Cars and the CUB birds dataset, both in an unconditional setting and in settings where the model is conditioned on class labels, attributes, and text. Finally, we propose an evaluation methodology that assesses the mesh and texture quality separately.

71.DeepRhythm: Exposing DeepFakes with Attentional Visual Heartbeat Rhythms ⬇️

As the GAN-based face image and video generation techniques, widely known as DeepFakes, have become more and more matured and realistic, the need for an effective DeepFakes detector has become imperative. Motivated by the fact that remote visualphotoplethysmography (PPG) is made possible by monitoring the minuscule periodic changes of skin color due to blood pumping through the face, we conjecture that normal heartbeat rhythms found in the real face videos will be diminished or even disrupted entirely in a DeepFake video, making it a powerful indicator for detecting DeepFakes. In this work, we show that our conjecture holds true and the proposed method indeed can very effectively exposeDeepFakes by monitoring the heartbeat rhythms, which is termedasDeepRhythm. DeepRhythm utilizes dual-spatial-temporal attention to adapt to dynamically changing face and fake types. Extensive experiments on FaceForensics++ and DFDC-preview datasets have demonstrated not only the effectiveness of our proposed method, but also how it can generalize over different datasets with various DeepFakes generation techniques and multifarious challenging degradations.

72.Equivariant Neural Rendering ⬇️

We propose a framework for learning neural scene representations directly from images, without 3D supervision. Our key insight is that 3D structure can be imposed by ensuring that the learned representation transforms like a real 3D scene. Specifically, we introduce a loss which enforces equivariance of the scene representation with respect to 3D transformations. Our formulation allows us to infer and render scenes in real time while achieving comparable results to models requiring minutes for inference. In addition, we introduce two challenging new datasets for scene representation and neural rendering, including scenes with complex lighting and backgrounds. Through experiments, we show that our model achieves compelling results on these datasets as well as on standard ShapeNet benchmarks.

73.DTG-Net: Differentiated Teachers Guided Self-Supervised Video Action Recognition ⬇️

State-of-the-art video action recognition models with complex network architecture have archived significant improvements, but these models heavily depend on large-scale well-labeled datasets. To reduce such dependency, we propose a self-supervised teacher-student architecture, i.e., the Differentiated Teachers Guided self-supervised Network (DTG-Net). In DTG-Net, except for reducing labeled data dependency by self-supervised learning (SSL), pre-trained action related models are used as teacher guidance providing prior knowledge to alleviate the demand for a large number of unlabeled videos in SSL. Specifically, leveraging the years of effort in action-related tasks, e.g., image classification, image-based action recognition, the DTG-Net learns the self-supervised video representation under various teacher guidance, i.e., those well-trained models of action-related tasks. Meanwhile, the DTG-Net is optimized in the way of contrastive self-supervised learning. When two image sequences are randomly sampled from the same video or different videos as the positive or negative pairs, respectively, they are then sent to the teacher and student networks for feature embedding. After that, the contrastive feature consistency is defined between features embedding of each pair, i.e., consistent for positive pair and inconsistent for negative pairs. Meanwhile, to reflect various teacher tasks' different guidance, we also explore different weighted guidance on teacher tasks. Finally, the DTG-Net is evaluated in two ways: (i) the self-supervised DTG-Net to pre-train the supervised action recognition models with only unlabeled videos; (ii) the supervised DTG-Net to be jointly trained with the supervised action networks in an end-to-end way. Its performance is better than most pre-training methods but also has excellent competitiveness compared to supervised action recognition methods.

74.HRDNet: High-resolution Detection Network for Small Objects ⬇️

Small object detection is challenging because small objects do not contain detailed information and may even disappear in the deep network. Usually, feeding high-resolution images into a network can alleviate this issue. However, simply enlarging the resolution will cause more problems, such as that, it aggravates the large variant of object scale and introduces unbearable computation cost. To keep the benefits of high-resolution images without bringing up new problems, we proposed the High-Resolution Detection Network (HRDNet). HRDNet takes multiple resolution inputs using multi-depth backbones. To fully take advantage of multiple features, we proposed Multi-Depth Image Pyramid Network (MD-IPN) and Multi-Scale Feature Pyramid Network (MS-FPN) in HRDNet. MD-IPN maintains multiple position information using multiple depth backbones. Specifically, high-resolution input will be fed into a shallow network to reserve more positional information and reducing the computational cost while low-resolution input will be fed into a deep network to extract more semantics. By extracting various features from high to low resolutions, the MD-IPN is able to improve the performance of small object detection as well as maintaining the performance of middle and large objects. MS-FPN is proposed to align and fuse multi-scale feature groups generated by MD-IPN to reduce the information imbalance between these multi-scale multi-level features. Extensive experiments and ablation studies are conducted on the standard benchmark dataset MS COCO2017, Pascal VOC2007/2012 and a typical small object dataset, VisDrone 2019. Notably, our proposed HRDNet achieves the state-of-the-art on these datasets and it performs better on small objects.

75.Faces à la Carte: Text-to-Face Generation via Attribute Disentanglement ⬇️

Text-to-Face (TTF) synthesis is a challenging task with great potential for diverse computer vision applications. Compared to Text-to-Image (TTI) synthesis tasks, the textual description of faces can be much more complicated and detailed due to the variety of facial attributes and the parsing of high dimensional abstract natural language. In this paper, we propose a Text-to-Face model that not only produces images in high resolution (1024x1024) with text-to-image consistency, but also outputs multiple diverse faces to cover a wide range of unspecified facial features in a natural way. By fine-tuning the multi-label classifier and image encoder, our model obtains the vectors and image embeddings which are used to transform the input noise vector sampled from the normal distribution. Afterwards, the transformed noise vector is fed into a pre-trained high-resolution image generator to produce a set of faces with the desired facial attributes. We refer to our model as TTF-HD. Experimental results show that TTF-HD generates high-quality faces with state-of-the-art performance.

76.Dynamic gesture retrieval: searching videos by human pose sequence ⬇️

The number of static human poses is limited, it is hard to retrieve the exact videos using one single pose as the clue. However, with a pose sequence or a dynamic gesture as the keyword, retrieving specific videos becomes more feasible. We propose a novel method for querying videos containing a designated sequence of human poses, whereas previous works only designate a single static pose. The proposed method takes continuous 3d human poses from keyword gesture video and video candidates, then converts each pose in individual frames into bone direction descriptors, which describe the direction of each natural connection in articulated pose. A temporal pyramid sliding window is then applied to find matches between designated gesture and video candidates, which ensures that same gestures with different duration can be matched.

77.NoPeopleAllowed: The Three-Step Approach to Weakly Supervised Semantic Segmentation ⬇️

We propose a novel approach to weakly supervised semantic segmentation, which consists of three consecutive steps. The first two steps extract high-quality pseudo masks from image-level annotated data, which are then used to train a segmentation model on the third step. The presented approach also addresses two problems in the data: class imbalance and missing labels. Using only image-level annotations as supervision, our method is capable of segmenting various classes and complex objects. It achieves 37.34 mean IoU on the test set, placing 3rd at the LID Challenge in the task of weakly supervised semantic segmentation.

78.Attribute-aware Identity-hard Triplet Loss for Video-based Person Re-identification ⬇️

Video-based person re-identification (Re-ID) is an important computer vision task. The batch-hard triplet loss frequently used in video-based person Re-ID suffers from the Distance Variance among Different Positives (DVDP) problem. In this paper, we address this issue by introducing a new metric learning method called Attribute-aware Identity-hard Triplet Loss (AITL), which reduces the intra-class variation among positive samples via calculating attribute distance. To achieve a complete model of video-based person Re-ID, a multi-task framework with Attribute-driven Spatio-Temporal Attention (ASTA) mechanism is also proposed. Extensive experiments on MARS and DukeMTMC-VID datasets shows that both the AITL and ASTA are very effective. Enhanced by them, even a simple light-weighted video-based person Re-ID baseline can outperform existing state-of-the-art approaches. The codes has been published on this https URL.

79.Semantic-driven Colorization ⬇️

Recent deep colorization works predict the semantic information implicitly while learning to colorize black-and-white photographic images. As a consequence, the generated color is easier to be overflowed, and the semantic faults are invisible. As human experience in coloring, the human first recognize which objects and their location in the photo, imagine which color is plausible for the objects as in real life, then colorize it. In this study, we simulate that human-like action to firstly let our network learn to segment what is in the photo, then colorize it. Therefore, our network can choose a plausible color under semantic constraint for specific objects, and give discriminative colors between them. Moreover, the segmentation map becomes understandable and interactable for the user. Our models are trained on PASCAL-Context and evaluated on selected images from the public domain and COCO-Stuff, which has several unseen categories compared to training data. As seen from the experimental results, our colorization system can provide plausible colors for specific objects and generate harmonious colors competitive with state-of-the-art methods.

80.Learning from the Scene and Borrowing from the Rich: Tackling the Long Tail in Scene Graph Generation ⬇️

Despite the huge progress in scene graph generation in recent years, its long-tail distribution in object relationships remains a challenging and pestering issue. Existing methods largely rely on either external knowledge or statistical bias information to alleviate this problem. In this paper, we tackle this issue from another two aspects: (1) scene-object interaction aiming at learning specific knowledge from a scene via an additive attention mechanism; and (2) long-tail knowledge transfer which tries to transfer the rich knowledge learned from the head into the tail. Extensive experiments on the benchmark dataset Visual Genome on three tasks demonstrate that our method outperforms current state-of-the-art competitors.

81.Mitigating Face Recognition Bias via Group Adaptive Classifier ⬇️

Face recognition is known to exhibit bias - subjects in certain demographic group can be better recognized than other groups. This work aims to learn a fair face representation, where faces of every group could be equally well-represented. Our proposed group adaptive classifier, GAC, learns to mitigate bias by using adaptive convolution kernels and attention mechanisms on faces based on their demographic attributes. The adaptive module comprises kernel masks and channel-wise attention maps for each demographic group so as to activate different facial regions for identification, leading to more discriminative features pertinent to their demographics. We also introduce an automated adaptation strategy which determines whether to apply adaptation to a certain layer by iteratively computing the dissimilarity among demographic-adaptive parameters, thereby increasing the efficiency of the adaptation learning. Experiments on benchmark face datasets (RFW, LFW, IJB-A, and IJB-C) show that our framework is able to mitigate face recognition bias on various demographic groups as well as maintain the competitive performance.

82.Unbiased Auxiliary Classifier GANs with MINE ⬇️

Auxiliary Classifier GANs (AC-GANs) are widely used conditional generative models and are capable of generating high-quality images. Previous work has pointed out that AC-GAN learns a biased distribution. To remedy this, Twin Auxiliary Classifier GAN (TAC-GAN) introduces a twin classifier to the min-max game. However, it has been reported that using a twin auxiliary classifier may cause instability in training. To this end, we propose an Unbiased Auxiliary GANs (UAC-GAN) that utilizes the Mutual Information Neural Estimator (MINE) to estimate the mutual information between the generated data distribution and labels. To further improve the performance, we also propose a novel projection-based statistics network architecture for MINE. Experimental results on three datasets, including Mixture of Gaussian (MoG), MNIST and CIFAR10 datasets, show that our UAC-GAN performs better than AC-GAN and TAC-GAN. Code can be found on the project website.

83.Accurate Anchor Free Tracking ⬇️

Visual object tracking is an important application of computer vision. Recently, Siamese based trackers have achieved good accuracy. However, most of Siamese based trackers are not efficient, as they exhaustively search potential object locations to define anchors and then classify each anchor (i.e., a bounding box). This paper develops the first Anchor Free Siamese Network (AFSN). Specifically, a target object is defined by a bounding box center, tracking offset, and object size. All three are regressed by Siamese network with no additional classification or regional proposal, and performed once for each frame. We also tune the stride and receptive field for Siamese network, and further perform ablation experiments to quantitatively illustrate the effectiveness of our AFSN. We evaluate AFSN using five most commonly used benchmarks and compare to the best anchor-based trackers with source codes available for each benchmark. AFSN is 3-425 times faster than these best anchor based trackers. AFSN is also 5.97% to 12.4% more accurate in terms of all metrics for benchmark sets OTB2015, VOT2015, VOT2016, VOT2018 and TrackingNet, except that SiamRPN++ is 4% better than AFSN in terms of Expected Average Overlap (EAO) on VOT2018 (but SiamRPN++ is 3.9 times slower).

84.GAN Memory with No Forgetting ⬇️

Seeking to address the fundamental issue of memory in lifelong learning, we propose a GAN memory that is capable of realistically remembering a stream of generative processes with \emph{no} forgetting. Our GAN memory is based on recognizing that one can modulate the ``style'' of a GAN model to form perceptually-distant targeted generation. Accordingly, we propose to do sequential style modulations atop a well-behaved base GAN model, to form sequential targeted generative models, while simultaneously benefiting from the transferred base knowledge. Experiments demonstrate the superiority of our method over existing approaches and its effectiveness in alleviating catastrophic forgetting for lifelong classification problems.

85.FakePolisher: Making DeepFakes More Detection-Evasive by Shallow Reconstruction ⬇️

The recently rapid advances of generative adversarial networks (GANs) in synthesizing realistic and natural DeepFake information (e.g., images, video) cause severe concerns and threats to our society. At this moment, GAN-based image generation methods are still imperfect, whose upsampling design has limitations in leaving some certain artifact patterns in the synthesized image. Such artifact patterns can be easily exploited (by recent methods) for difference detection of real and GAN-synthesized images.
To reduce the artifacts in the synthesized images, deep reconstruction techniques are usually futile because the process itself can leave traces of artifacts. In this paper, we devise a simple yet powerful approach termed FakePolisher that performs shallow reconstruction of fake images through learned linear dictionary, intending to effectively and efficiently reduce the artifacts introduced during image synthesis. The comprehensive evaluation on 3 state-of-the-art DeepFake detection methods and fake images generated by 16 popular GAN-based fake image generation techniques, demonstrates the effectiveness of our technique.

86.CBR-Net: Cascade Boundary Refinement Network for Action Detection: Submission to ActivityNet Challenge 2020 (Task 1) ⬇️

In this report, we present our solution for the task of temporal action localization (detection) (task 1) in ActivityNet Challenge 2020. The purpose of this task is to temporally localize intervals where actions of interest occur and predict the action categories in a long untrimmed video. Our solution mainly includes three components: 1) feature encoding: we apply three kinds of backbones, including TSN [7], Slowfast[3] and I3d[1], which are both pretrained on Kinetics dataset[2]. Applying these models, we can extract snippet-level video representations; 2) proposal generation: we choose BMN [5] as our baseline, base on which we design a Cascade Boundary Refinement Network (CBR-Net) to conduct proposal detection. The CBR-Net mainly contains two modules: temporal feature encoding, which applies BiLSTM to encode long-term temporal information; CBR module, which targets to refine the proposal precision under different parameter settings; 3) action localization: In this stage, we combine the video-level classification results obtained by the fine tuning networks to predict the category of each proposal. Moreover, we also apply to different ensemble strategies to improve the performance of the designed solution, by which we achieve 42.788% on the testing set of ActivityNet v1.3 dataset in terms of mean Average Precision metrics.

87.Self-Supervised Discovery of Anatomical Shape Landmarks ⬇️

Statistical shape analysis is a very useful tool in a wide range of medical and biological applications. However, it typically relies on the ability to produce a relatively small number of features that can capture the relevant variability in a population. State-of-the-art methods for obtaining such anatomical features rely on either extensive preprocessing or segmentation and/or significant tuning and post-processing. These shortcomings limit the widespread use of shape statistics. We propose that effective shape representations should provide sufficient information to align/register images. Using this assumption we propose a self-supervised, neural network approach for automatically positioning and detecting landmarks in images that can be used for subsequent analysis. The network discovers the landmarks corresponding to anatomical shape features that promote good image registration in the context of a particular class of transformations. In addition, we also propose a regularization for the proposed network which allows for a uniform distribution of these discovered landmarks. In this paper, we present a complete framework, which only takes a set of input images and produces landmarks that are immediately usable for statistical shape analysis. We evaluate the performance on a phantom dataset as well as 2D and 3D images.

88.Temporal Fusion Network for Temporal Action Localization:Submission to ActivityNet Challenge 2020 (Task E) ⬇️

This technical report analyzes a temporal action localization method we used in the HACS competition which is hosted in Activitynet Challenge 2020.The goal of our task is to locate the start time and end time of the action in the untrimmed video, and predict action category.Firstly, we utilize the video-level feature information to train multiple video-level action classification models. In this way, we can get the category of action in the video.Secondly, we focus on generating high quality temporal proposals.For this purpose, we apply BMN to generate a large number of proposals to obtain high recall rates. We then refine these proposals by employing a cascade structure network called Refine Network, which can predict position offset and new IOU under the supervision of ground this http URL make the proposals more accurate, we use bidirectional LSTM, Nonlocal and Transformer to capture temporal relationships between local features of each proposal and global features of the video data.Finally, by fusing the results of multiple models, our method obtains 40.55% on the validation set and 40.53% on the test set in terms of mAP, and achieves Rank 1 in this challenge.

89.Weakly-supervised Any-shot Object Detection ⬇️

Methods for object detection and segmentation rely on large scale instance-level annotations for training, which are difficult and time-consuming to collect. Efforts to alleviate this look at varying degrees and quality of supervision. Weakly-supervised approaches draw on image-level labels to build detectors/segmentors, while zero/few-shot methods assume abundant instance-level data for a set of base classes, and none to a few examples for novel classes. This taxonomy has largely siloed algorithmic designs. In this work, we aim to bridge this divide by proposing an intuitive weakly-supervised model that is applicable to a range of supervision: from zero to a few instance-level samples per novel class. For base classes, our model learns a mapping from weakly-supervised to fully-supervised detectors/segmentors. By learning and leveraging visual and lingual similarities between the novel and base classes, we transfer those mappings to obtain detectors/segmentors for novel classes; refining them with a few novel class instance-level annotated samples, if available. The overall model is end-to-end trainable and highly flexible. Through extensive experiments on MS-COCO and Pascal VOC benchmark datasets we show improved performance in a variety of settings.

90.Multi-Modal Fingerprint Presentation Attack Detection: Evaluation On A New Dataset ⬇️

Fingerprint presentation attack detection is becoming an increasingly challenging problem due to the continuous advancement of attack preparation techniques, which generate realistic-looking fake fingerprint presentations. In this work, rather than relying on legacy fingerprint images, which are widely used in the community, we study the usefulness of multiple recently introduced sensing modalities. Our study covers front-illumination imaging using short-wave-infrared, near-infrared, and laser illumination; and back-illumination imaging using near-infrared light. Toward studying the effectiveness of each of these unconventional sensing modalities and their fusion for liveness detection, we conducted a comprehensive analysis using a fully convolutional deep neural network framework. Our evaluation compares different combination of the new sensing modalities to legacy data from one of our collections as well as the public LivDet2015 dataset, showing the superiority of the new sensing modalities in most cases. It also covers the cases of known and unknown attacks and the cases of intra-dataset and inter-dataset evaluations. Our results indicate that the power of our approach stems from the nature of the captured data rather than the employed classification framework, which justifies the extra cost for hardware-based (or hybrid) solutions. We plan to publicly release one of our dataset collections.

91.OrigamiNet: Weakly-Supervised, Segmentation-Free, One-Step, Full Page Text Recognition by learning to unfold ⬇️

Text recognition is a major computer vision task with a big set of associated challenges. One of those traditional challenges is the coupled nature of text recognition and segmentation. This problem has been progressively solved over the past decades, going from segmentation based recognition to segmentation free approaches, which proved more accurate and much cheaper to annotate data for. We take a step from segmentation-free single line recognition towards segmentation-free multi-line / full page recognition. We propose a novel and simple neural network module, termed \textbf{OrigamiNet}, that can augment any CTC-trained, fully convolutional single line text recognizer, to convert it into a multi-line version by providing the model with enough spatial capacity to be able to properly collapse a 2D input signal into 1D without losing information. Such modified networks can be trained using exactly their same simple original procedure, and using only \textbf{unsegmented} image and text pairs. We carry out a set of interpretability experiments that show that our trained models learn an accurate implicit line segmentation. We achieve state-of-the-art character error rate on both IAM & ICDAR 2017 HTR benchmarks for handwriting recognition, surpassing all other methods in the literature. On IAM we even surpass single line methods that use accurate localization information during training. Our code is available online at \url{this https URL}.

92.Multispectral Biometrics System Framework: Application to Presentation Attack Detection ⬇️

In this work, we present a general framework for building a biometrics system capable of capturing multispectral data from a series of sensors synchronized with active illumination sources. The framework unifies the system design for different biometric modalities and its realization on face, finger and iris data is described in detail. To the best of our knowledge, the presented design is the first to employ such a diverse set of electromagnetic spectrum bands, ranging from visible to long-wave-infrared wavelengths, and is capable of acquiring large volumes of data in seconds. Having performed a series of data collections, we run a comprehensive analysis on the captured data using a deep-learning classifier for presentation attack detection. Our study follows a data-centric approach attempting to highlight the strengths and weaknesses of each spectral band at distinguishing live from fake samples.

93.Early Blindness Detection Based on Retinal Images Using Ensemble Learning ⬇️

Diabetic retinopathy (DR) is the primary cause of vision loss among grownup people around the world. In four out of five cases having diabetes for a prolonged period leads to DR. If detected early, more than 90 percent of the new DR occurrences can be prevented from turning into blindness through proper treatment. Despite having multiple treatment procedures available that are well capable to deal with DR, the negligence and failure of early detection cost most of the DR patients their precious eyesight. The recent developments in the field of Digital Image Processing (DIP) and Machine Learning (ML) have paved the way to use machines in this regard. The contemporary technologies allow us to develop devices capable of automatically detecting the condition of a persons eyes based on their retinal images. However, in practice, several factors hinder the quality of the captured images and impede the detection outcome. In this study, a novel early blind detection method has been proposed based on the color information extracted from retinal images using an ensemble learning algorithm. The method has been tested on a set of retinal images collected from people living in the rural areas of South Asia, which resulted in a 91 percent classification accuracy.

94.Learning-to-Learn Personalised Human Activity Recognition Models ⬇️

Human Activity Recognition~(HAR) is the classification of human movement, captured using one or more sensors either as wearables or embedded in the environment~(e.g. depth cameras, pressure mats). State-of-the-art methods of HAR rely on having access to a considerable amount of labelled data to train deep architectures with many train-able parameters. This becomes prohibitive when tasked with creating models that are sensitive to personal nuances in human movement, explicitly present when performing exercises. In addition, it is not possible to collect training data to cover all possible subjects in the target population. Accordingly, learning personalised models with few data remains an interesting challenge for HAR research. We present a meta-learning methodology for learning to learn personalised HAR models for HAR; with the expectation that the end-user need only provides a few labelled data but can benefit from the rapid adaptation of a generic meta-model. We introduce two algorithms, Personalised MAML and Personalised Relation Networks inspired by existing Meta-Learning algorithms but optimised for learning HAR models that are adaptable to any person in health and well-being applications. A comparative study shows significant performance improvements against the state-of-the-art Deep Learning algorithms and the Few-shot Meta-Learning algorithms in multiple HAR domains.

95.Defending against GAN-based Deepfake Attacks via Transformation-aware Adversarial Faces ⬇️

Deepfake represents a category of face-swapping attacks that leverage machine learning models such as autoencoders or generative adversarial networks. Although the concept of the face-swapping is not new, its recent technical advances make fake content (e.g., images, videos) more realistic and imperceptible to Humans. Various detection techniques for Deepfake attacks have been explored. These methods, however, are passive measures against Deepfakes as they are mitigation strategies after the high-quality fake content is generated. More importantly, we would like to think ahead of the attackers with robust defenses. This work aims to take an offensive measure to impede the generation of high-quality fake images or videos. Specifically, we propose to use novel transformation-aware adversarially perturbed faces as a defense against GAN-based Deepfake attacks. Different from the naive adversarial faces, our proposed approach leverages differentiable random image transformations during the generation. We also propose to use an ensemble-based approach to enhance the defense robustness against GAN-based Deepfake variants under the black-box setting. We show that training a Deepfake model with adversarial faces can lead to a significant degradation in the quality of synthesized faces. This degradation is twofold. On the one hand, the quality of the synthesized faces is reduced with more visual artifacts such that the synthesized faces are more obviously fake or less convincing to human observers. On the other hand, the synthesized faces can easily be detected based on various metrics.

96.The DeepFake Detection Challenge Dataset ⬇️

Deepfakes are a recent off-the-shelf manipulation technique that allows anyone to swap two identities in a single video. In addition to Deepfakes, a variety of GAN-based face swapping methods have also been published with accompanying code. To counter this emerging threat, we have constructed an extremely large face swap video dataset to enable the training of detection models, and organized the accompanying DeepFake Detection Challenge (DFDC) Kaggle competition. Importantly, all recorded subjects agreed to participate in and have their likenesses modified during the construction of the face-swapped dataset. The DFDC dataset is by far the largest currently and publicly available face swap video dataset, with over 100,000 total clips sourced from 3,426 paid actors, produced with several Deepfake, GAN-based, and non-learned methods. In addition to describing the methods used to construct the dataset, we provide a detailed analysis of the top submissions from the Kaggle contest. We show although Deepfake detection is extremely difficult and still an unsolved problem, a Deepfake detection model trained only on the DFDC can generalize to real "in-the-wild" Deepfake videos, and such a model can be a valuable analysis tool when analyzing potentially Deepfaked videos. Training, validation and testing corpuses can be downloaded from this http URL (URL to be updated).

97.Spectral DiffuserCam: lensless snapshot hyperspectral imaging with a spectral filter array ⬇️

Hyperspectral imaging is useful for applications ranging from medical diagnostics to crop monitoring; however, traditional scanning hyperspectral imagers are prohibitively slow and expensive for widespread adoption. Snapshot techniques exist but are often confined to bulky benchtop setups or have low spatio-spectral resolution. In this paper, we propose a novel, compact, and inexpensive computational camera for snapshot hyperspectral imaging. Our system consists of a repeated spectral filter array placed directly on the image sensor and a diffuser placed close to the sensor. Each point in the world maps to a unique pseudorandom pattern on the spectral filter array, which encodes multiplexed spatio-spectral information. A sparsity-constrained inverse problem solver then recovers the hyperspectral volume with good spatio-spectral resolution. By using a spectral filter array, our hyperspectral imaging framework is flexible and can be designed with contiguous or non-contiguous spectral filters that can be chosen for a given application. We provide theory for system design, demonstrate a prototype device, and present experimental results with high spatio-spectral resolution.

98.Learning Diverse and Discriminative Representations via the Principle of Maximal Coding Rate Reduction ⬇️

To learn intrinsic low-dimensional structures from high-dimensional data that most discriminate between classes, we propose the principle of Maximal Coding Rate Reduction ($\text{MCR}^2$), an information-theoretic measure that maximizes the coding rate difference between the whole dataset and the sum of each individual class. We clarify its relationships with most existing frameworks such as cross-entropy, information bottleneck, information gain, contractive and contrastive learning, and provide theoretical guarantees for learning diverse and discriminative features. The coding rate can be accurately computed from finite samples of degenerate subspace-like distributions and can learn intrinsic representations in supervised, self-supervised, and unsupervised settings in a unified manner. Empirically, the representations learned using this principle alone are significantly more robust to label corruptions in classification than those using cross-entropy, and can lead to state-of-the-art results in clustering mixed data from self-learned invariant features.

99.Efficient Black-Box Adversarial Attack Guided by the Distribution of Adversarial Perturbations ⬇️

This work studied the score-based black-box adversarial attack problem, where only a continuous score is returned for each query, while the structure and parameters of the attacked model are unknown. A promising approach to solve this problem is evolution strategies (ES), which introduces a search distribution to sample perturbations that are likely to be adversarial. Gaussian distribution is widely adopted as the search distribution in the standard ES algorithm. However, it may not be flexible enough to capture the diverse distributions of adversarial perturbations around different benign examples. In this work, we propose to transform the Gaussian-distributed variable to another space through a conditional flow-based model, to enhance the capability and flexibility of capturing the intrinsic distribution of adversarial perturbations conditioned on the benign example. Besides, to further enhance the query efficiency, we propose to pre-train the conditional flow model based on some white-box surrogate models, utilizing the transferability of adversarial perturbations across different models, which has been widely observed in the literature of adversarial examples. Consequently, the proposed method could take advantage of both query-based and transfer-based attack methods, to achieve satisfying attack performance on both effectiveness and efficiency. Extensive experiments of attacking four target models on CIFAR-10 and Tiny-ImageNet verify the superior performance of the proposed method to state-of-the-art methods.

100.Improved Conditional Flow Models for Molecule to Image Synthesis ⬇️

In this paper, we aim to synthesize cell microscopy images under different molecular interventions, motivated by practical applications to drug development. Building on the recent success of graph neural networks for learning molecular embeddings and flow-based models for image generation, we propose Mol2Image: a flow-based generative model for molecule to cell image synthesis. To generate cell features at different resolutions and scale to high-resolution images, we develop a novel multi-scale flow architecture based on a Haar wavelet image pyramid. To maximize the mutual information between the generated images and the molecular interventions, we devise a training strategy based on contrastive learning. To evaluate our model, we propose a new set of metrics for biological image generation that are robust, interpretable, and relevant to practitioners. We show quantitatively that our method learns a meaningful embedding of the molecular intervention, which is translated into an image representation reflecting the biological effects of the intervention.

101.The Limit of the Batch Size ⬇️

Large-batch training is an efficient approach for current distributed deep learning systems. It has enabled researchers to reduce the ImageNet/ResNet-50 training from 29 hours to around 1 minute. In this paper, we focus on studying the limit of the batch size. We think it may provide a guidance to AI supercomputer and algorithm designers. We provide detailed numerical optimization instructions for step-by-step comparison. Moreover, it is important to understand the generalization and optimization performance of huge batch training. Hoffer et al. introduced "ultra-slow diffusion" theory to large-batch training. However, our experiments show contradictory results with the conclusion of Hoffer et al. We provide comprehensive experimental results and detailed analysis to study the limitations of batch size scaling and "ultra-slow diffusion" theory. For the first time we scale the batch size on ImageNet to at least a magnitude larger than all previous work, and provide detailed studies on the performance of many state-of-the-art optimization schemes under this setting. We propose an optimization recipe that is able to improve the top-1 test accuracy by 18% compared to the baseline.

102.APQ: Joint Search for Network Architecture, Pruning and Quantization Policy ⬇️

We present APQ for efficient deep learning inference on resource-constrained hardware. Unlike previous methods that separately search the neural architecture, pruning policy, and quantization policy, we optimize them in a joint manner. To deal with the larger design space it brings, a promising approach is to train a quantization-aware accuracy predictor to quickly get the accuracy of the quantized model and feed it to the search engine to select the best fit. However, training this quantization-aware accuracy predictor requires collecting a large number of quantized <model, accuracy> pairs, which involves quantization-aware finetuning and thus is highly time-consuming. To tackle this challenge, we propose to transfer the knowledge from a full-precision (i.e., fp32) accuracy predictor to the quantization-aware (i.e., int8) accuracy predictor, which greatly improves the sample efficiency. Besides, collecting the dataset for the fp32 accuracy predictor only requires to evaluate neural networks without any training cost by sampling from a pretrained once-for-all network, which is highly efficient. Extensive experiments on ImageNet demonstrate the benefits of our joint optimization approach. With the same accuracy, APQ reduces the latency/energy by 2x/1.3x over MobileNetV2+HAQ. Compared to the separate optimization approach (ProxylessNAS+AMC+HAQ), APQ achieves 2.3% higher ImageNet accuracy while reducing orders of magnitude GPU hours and CO2 emission, pushing the frontier for green AI that is environmental-friendly. The code and video are publicly available.

103.Deep learning mediated single time-point image-based prediction of embryo developmental outcome at the cleavage stage ⬇️

In conventional clinical in-vitro fertilization practices embryos are transferred either at the cleavage or blastocyst stages of development. Cleavage stage transfers, particularly, are beneficial for patients with relatively poor prognosis and at fertility centers in resource-limited settings where there is a higher chance of developmental failure in embryos in-vitro. However, one of the major limitations of embryo selections at the cleavage stage is the availability of very low number of manually discernable features to predict developmental outcomes. Although, time-lapse imaging systems have been proposed as possible solutions, they are cost-prohibitive and require bulky and expensive hardware, and labor-intensive. Advances in convolutional neural networks (CNNs) have been utilized to provide accurate classifications across many medical and non-medical object categories. Here, we report an automated system for classification and selection of human embryos at the cleavage stage using a trained CNN combined with a genetic algorithm. The system selected the cleavage stage embryo at 70 hours post insemination (hpi) that ultimately developed into top-quality blastocyst at 70 hpi with 64% accuracy, outperforming the abilities of embryologists in identifying embryos with the highest developmental potential. Such systems can have a significant impact on IVF procedures by empowering embryologists for accurate and consistent embryo assessment in both resource-poor and resource-rich settings.

104.A Dataset and Benchmarks for Multimedia Social Analysis ⬇️

We present a new publicly available dataset with the goal of advancing multi-modality learning by offering vision and language data within the same context. This is achieved by obtaining data from a social media website with posts containing multiple paired images/videos and text, along with comment trees containing images/videos and/or text. With a total of 677k posts, 2.9 million post images, 488k post videos, 1.4 million comment images, 4.6 million comment videos, and 96.9 million comments, data from different modalities can be jointly used to improve performances for a variety of tasks such as image captioning, image classification, next frame prediction, sentiment analysis, and language modeling. We present a wide range of statistics for our dataset. Finally, we provide baseline performance analysis for one of the regression tasks using pre-trained models and several fully connected networks.

105.Differentiable Neural Architecture Transformation for Reproducible Architecture Improvement ⬇️

Recently, Neural Architecture Search (NAS) methods are introduced and show impressive performance on many benchmarks. Among those NAS studies, Neural Architecture Transformer (NAT) aims to improve the given neural architecture to have better performance while maintaining computational costs. However, NAT has limitations about a lack of reproducibility. In this paper, we propose differentiable neural architecture transformation that is reproducible and efficient. The proposed method shows stable performance on various architectures. Extensive reproducibility experiments on two datasets, i.e., CIFAR-10 and Tiny Imagenet, present that the proposed method definitely outperforms NAT and be applicable to other models and datasets.

106.Slowing Down the Weight Norm Increase in Momentum-based Optimizers ⬇️

Normalization techniques, such as batch normalization (BN), have led to significant improvements in deep neural network performances. Prior studies have analyzed the benefits of the resulting scale invariance of the weights for the gradient descent (GD) optimizers: it leads to a stabilized training due to the auto-tuning of step sizes. However, we show that, combined with the momentum-based algorithms, the scale invariance tends to induce an excessive growth of the weight norms. This in turn overly suppresses the effective step sizes during training, potentially leading to sub-optimal performances in deep neural networks. We analyze this phenomenon both theoretically and empirically. We propose a simple and effective solution: at each iteration of momentum-based GD optimizers (e.g. SGD or Adam) applied on scale-invariant weights (e.g. Conv weights preceding a BN layer), we remove the radial component (i.e. parallel to the weight vector) from the update vector. Intuitively, this operation prevents the unnecessary update along the radial direction that only increases the weight norm without contributing to the loss minimization. We verify that the modified optimizers SGDP and AdamP successfully regularize the norm growth and improve the performance of a broad set of models. Our experiments cover tasks including image classification and retrieval, object detection, robustness benchmarks, and audio classification. Source code is available at this https URL.

107.Dissimilarity Mixture Autoencoder for Deep Clustering ⬇️

In this paper, we introduce the Dissimilarity Mixture Autoencoder (DMAE), a novel neural network model that uses a dissimilarity function to generalize a family of density estimation and clustering methods. It is formulated in such a way that it internally estimates the parameters of a probability distribution through gradient-based optimization. Also, the proposed model can leverage from deep representation learning due to its straightforward incorporation into deep learning architectures, because, it consists of an encoder-decoder network that computes a probabilistic representation. Experimental evaluation was performed on image and text clustering benchmark datasets showing that the method is competitive in terms of unsupervised classification accuracy and normalized mutual information. The source code to replicate the experiments is publicly available at this https URL

108.Emotion Recognition in Audio and Video Using Deep Neural Networks ⬇️

Humans are able to comprehend information from multiple domains for e.g. speech, text and visual. With advancement of deep learning technology there has been significant improvement of speech recognition. Recognizing emotion from speech is important aspect and with deep learning technology emotion recognition has improved in accuracy and latency. There are still many challenges to improve accuracy. In this work, we attempt to explore different neural networks to improve accuracy of emotion recognition. With different architectures explored, we find (CNN+RNN) + 3DCNN multi-model architecture which processes audio spectrograms and corresponding video frames giving emotion prediction accuracy of 54.0% among 4 emotions and 71.75% among 3 emotions using IEMOCAP[2] dataset.

109.Generalized Adversarially Learned Inference ⬇️

Allowing effective inference of latent vectors while training GANs can greatly increase their applicability in various downstream tasks. Recent approaches, such as ALI and BiGAN frameworks, develop methods of inference of latent variables in GANs by adversarially training an image generator along with an encoder to match two joint distributions of image and latent vector pairs. We generalize these approaches to incorporate multiple layers of feedback on reconstructions, self-supervision, and other forms of supervision based on prior or learned knowledge about the desired solutions. We achieve this by modifying the discriminator's objective to correctly identify more than two joint distributions of tuples of an arbitrary number of random variables consisting of images, latent vectors, and other variables generated through auxiliary tasks, such as reconstruction and inpainting or as outputs of suitable pre-trained models. We design a non-saturating maximization objective for the generator-encoder pair and prove that the resulting adversarial game corresponds to a global optimum that simultaneously matches all the distributions. Within our proposed framework, we introduce a novel set of techniques for providing self-supervised feedback to the model based on properties, such as patch-level correspondence and cycle consistency of reconstructions. Through comprehensive experiments, we demonstrate the efficacy, scalability, and flexibility of the proposed approach for a variety of tasks.

110.CompressNet: Generative Compression at Extremely Low Bitrates ⬇️

Compressing images at extremely low bitrates (< 0.1 bpp) has always been a challenging task since the quality of reconstruction significantly reduces due to the strong imposed constraint on the number of bits allocated for the compressed data. With the increasing need to transfer large amounts of images with limited bandwidth, compressing images to very low sizes is a crucial task. However, the existing methods are not effective at extremely low bitrates. To address this need, we propose a novel network called CompressNet which augments a Stacked Autoencoder with a Switch Prediction Network (SAE-SPN). This helps in the reconstruction of visually pleasing images at these low bitrates (< 0.1 bpp). We benchmark the performance of our proposed method on the Cityscapes dataset, evaluating over different metrics at extremely low bitrates to show that our method outperforms the other state-of-the-art. In particular, at a bitrate of 0.07, CompressNet achieves 22% lower Perceptual Loss and 55% lower Frechet Inception Distance (FID) compared to the deep learning SOTA methods.

111.Leveraging Multimodal Behavioral Analytics for Automated Job Interview Performance Assessment and Feedback ⬇️

Behavioral cues play a significant part in human communication and cognitive perception. In most professional domains, employee recruitment policies are framed such that both professional skills and personality traits are adequately assessed. Hiring interviews are structured to evaluate expansively a potential employee's suitability for the position - their professional qualifications, interpersonal skills, ability to perform in critical and stressful situations, in the presence of time and resource constraints, etc. Therefore, candidates need to be aware of their positive and negative attributes and be mindful of behavioral cues that might have adverse effects on their success. We propose a multimodal analytical framework that analyzes the candidate in an interview scenario and provides feedback for predefined labels such as engagement, speaking rate, eye contact, etc. We perform a comprehensive analysis that includes the interviewee's facial expressions, speech, and prosodic information, using the video, audio, and text transcripts obtained from the recorded interview. We use these multimodal data sources to construct a composite representation, which is used for training machine learning classifiers to predict the class labels. Such analysis is then used to provide constructive feedback to the interviewee for their behavioral cues and body language. Experimental validation showed that the proposed methodology achieved promising results.

112.Continual General Chunking Problem and SyncMap ⬇️

Humans possess an inherent ability to chunk sequences into their constituent parts. In fact, this ability is thought to bootstrap language skills to the learning of image patterns which might be a key to a more animal-like type of intelligence. Here, we propose a continual generalization of the chunking problem (an unsupervised problem), encompassing fixed and probabilistic chunks, discovery of temporal and causal structures and their continual variations. Additionally, we propose an algorithm called SyncMap that can learn and adapt to changes in the problem by creating a dynamic map which preserves the correlation between variables. Results of SyncMap suggest that the proposed algorithm learn near optimal solutions, despite the presence of many types of structures and their continual variation. When compared to Word2vec, PARSER and MRIL, SyncMap surpasses or ties with the best algorithm on $77%$ of the scenarios while being the second best in the remaing $23%$.

113.Structural Autoencoders Improve Representations for Generation and Transfer ⬇️

We study the problem of structuring a learned representation to significantly improve performance without supervision. Unlike most methods which focus on using side information like weak supervision or defining new regularization objectives, we focus on improving the learned representation by structuring the architecture of the model. We propose a self-attention based architecture to make the encoder explicitly associate parts of the representation with parts of the input observation. Meanwhile, our structural decoder architecture encourages a hierarchical structure in the latent space, akin to structural causal models, and learns a natural ordering of the latent mechanisms. We demonstrate how these models learn a representation which improves results in a variety of downstream tasks including generation, disentanglement, and transfer using several challenging and natural image datasets.

114.DeeperGCN: All You Need to Train Deeper GCNs ⬇️

Graph Convolutional Networks (GCNs) have been drawing significant attention with the power of representation learning on graphs. Unlike Convolutional Neural Networks (CNNs), which are able to take advantage of stacking very deep layers, GCNs suffer from vanishing gradient, over-smoothing and over-fitting issues when going deeper. These challenges limit the representation power of GCNs on large-scale graphs. This paper proposes DeeperGCN that is capable of successfully and reliably training very deep GCNs. We define differentiable generalized aggregation functions to unify different message aggregation operations (e.g. mean, max). We also propose a novel normalization layer namely MsgNorm and a pre-activation version of residual connections for GCNs. Extensive experiments on Open Graph Benchmark (OGB) show DeeperGCN significantly boosts performance over the state-of-the-art on the large scale graph learning tasks of node property prediction and graph property prediction. Please visit this https URL for more information.

115.Bootstrap Your Own Latent: A New Approach to Self-Supervised Learning ⬇️

We introduce Bootstrap Your Own Latent (BYOL), a new approach to self-supervised image representation learning. BYOL relies on two neural networks, referred to as online and target networks, that interact and learn from each other. From an augmented view of an image, we train the online network to predict the target network representation of the same image under a different augmented view. At the same time, we update the target network with a slow-moving average of the online network. While state-of-the art methods intrinsically rely on negative pairs, BYOL achieves a new state of the art without them. BYOL reaches $74.3%$ top-1 classification accuracy on ImageNet using the standard linear evaluation protocol with a ResNet-50 architecture and $79.6%$ with a larger ResNet. We show that BYOL performs on par or better than the current state of the art on both transfer and semi-supervised benchmarks.

116.RoadNet-RT: High Throughput CNN Architecture and SoC Design for Real-Time Road Segmentation ⬇️

In recent years, convolutional neural network has gained popularity in many engineering applications especially for computer vision. In order to achieve better performance, often more complex structures and advanced operations are incorporated into the neural networks, which results very long inference time. For time-critical tasks such as autonomous driving and virtual reality, real-time processing is fundamental. In order to reach real-time process speed, a light-weight, high-throughput CNN architecture namely RoadNet-RT is proposed for road segmentation in this paper. It achieves 90.33% MaxF score on test set of KITTI road segmentation task and 8 ms per frame when running on GTX 1080 GPU. Comparing to the state-of-the-art network, RoadNet-RT speeds up the inference time by a factor of 20 at the cost of only 6.2% accuracy loss. For hardware design optimization, several techniques such as depthwise separable convolution and non-uniformed kernel size convolution are customized designed to further reduce the processing time. The proposed CNN architecture has been successfully implemented on an FPGA ZCU102 MPSoC platform that achieves the computation capability of 83.05 GOPS. The system throughput reaches 327.9 frames per second with image size 1216x176.

117.Adversarial Self-Supervised Contrastive Learning ⬇️

Existing adversarial learning approaches mostly use class labels to generate adversarial samples that lead to incorrect predictions, which are then used to augment the training of the model for improved robustness. While some recent works propose semi-supervised adversarial learning methods that utilize unlabeled data, they still require class labels. However, do we really need class labels at all, for adversarially robust training of deep neural networks? In this paper, we propose a novel adversarial attack for unlabeled data, which makes the model confuse the instance-level identities of the perturbed data samples. Further, we present a self-supervised contrastive learning framework to adversarially train a robust neural network without labeled data, which aims to maximize the similarity between a random augmentation of a data sample and its instance-wise adversarial perturbation. We validate our method, Robust Contrastive Learning (RoCL), on multiple benchmark datasets, on which it obtains comparable robust accuracy over state-of-the-art supervised adversarial learning methods, and significantly improved robustness against the black box and unseen types of attacks. Moreover, with further joint fine-tuning with supervised adversarial loss, RoCL obtains even higher robust accuracy over using self-supervised learning alone. Notably, RoCL also demonstrate impressive results in robust transfer learning.

118.Sparse Separable Nonnegative Matrix Factorization ⬇️

We propose a new variant of nonnegative matrix factorization (NMF), combining separability and sparsity assumptions. Separability requires that the columns of the first NMF factor are equal to columns of the input matrix, while sparsity requires that the columns of the second NMF factor are sparse. We call this variant sparse separable NMF (SSNMF), which we prove to be NP-complete, as opposed to separable NMF which can be solved in polynomial time. The main motivation to consider this new model is to handle underdetermined blind source separation problems, such as multispectral image unmixing. We introduce an algorithm to solve SSNMF, based on the successive nonnegative projection algorithm (SNPA, an effective algorithm for separable NMF), and an exact sparse nonnegative least squares solver. We prove that, in noiseless settings and under mild assumptions, our algorithm recovers the true underlying sources. This is illustrated by experiments on synthetic data sets and the unmixing of a multispectral image.

119.Rethinking the Value of Labels for Improving Class-Imbalanced Learning ⬇️

Real-world data often exhibits long-tailed distributions with heavy class imbalance, posing great challenges for deep recognition models. We identify a persisting dilemma on the value of labels in the context of imbalanced learning: on the one hand, supervision from labels typically leads to better results than its unsupervised counterparts; on the other hand, heavily imbalanced data naturally incurs "label bias" in the classifier, where the decision boundary can be drastically altered by the majority classes. In this work, we systematically investigate these two facets of labels. We demonstrate, theoretically and empirically, that class-imbalanced learning can significantly benefit in both semi-supervised and self-supervised manners. Specifically, we confirm that (1) positively, imbalanced labels are valuable: given more unlabeled data, the original labels can be leveraged with the extra data to reduce label bias in a semi-supervised manner, which greatly improves the final classifier; (2) negatively however, we argue that imbalanced labels are not useful always: classifiers that are first pre-trained in a self-supervised manner consistently outperform their corresponding baselines. Extensive experiments on large-scale imbalanced datasets verify our theoretically grounded strategies, showing superior performance over the previous state-of-the-arts. Our intriguing findings highlight the need to rethink the usage of imbalanced labels in realistic long-tailed tasks.

120.TURB-Rot. A large database of 3d and 2d snapshots from turbulent rotating flows ⬇️

We present TURB-Rot, a new open database of 3d and 2d snapshots of turbulent velocity fields, obtained by Direct Numerical Simulations (DNS) of the original Navier-Stokes equations in the presence of rotation. The aim is to provide the community interested in data-assimilation and/or computer vision with a new testing-ground made of roughly 300K complex images and fields. TURB-Rot data are characterized by multi-scales strongly non-Gaussian features and rough, non-differentiable, fields over almost two decades of scales. In addition, coming from fully resolved numerical simulations of the original partial differential equations, they offer the possibility to apply a wide range of approaches, from equation-free to physics-based models. TURB-Rot data are reachable at this http URL

121.BI-MAML: Balanced Incremental Approach for Meta Learning ⬇️

We present a novel Balanced Incremental Model Agnostic Meta Learning system (BI-MAML) for learning multiple tasks. Our method implements a meta-update rule to incrementally adapt its model to new tasks without forgetting old tasks. Such a capability is not possible in current state-of-the-art MAML approaches. These methods effectively adapt to new tasks, however, suffer from 'catastrophic forgetting' phenomena, in which new tasks that are streamed into the model degrade the performance of the model on previously learned tasks. Our system performs the meta-updates with only a few-shots and can successfully accomplish them. Our key idea for achieving this is the design of balanced learning strategy for the baseline model. The strategy sets the baseline model to perform equally well on various tasks and incorporates time efficiency. The balanced learning strategy enables BI-MAML to both outperform other state-of-the-art models in terms of classification accuracy for existing tasks and also accomplish efficient adaption to similar new tasks with less required shots. We evaluate BI-MAML by conducting comparisons on two common benchmark datasets with multiple number of image classification tasks. BI-MAML performance demonstrates advantages in both accuracy and efficiency.