20201103.md

ArXiv cs.CV --Tue, 3 Nov 2020

1.Continuous and Diverse Image-to-Image Translation via Signed Attribute Vectors ⬇️

Recent image-to-image (I2I) translation algorithms focus on learning the mapping from a source to a target domain. However, the continuous translation problem that synthesizes intermediate results between the two domains has not been well-studied in the literature. Generating a smooth sequence of intermediate results bridges the gap of two different domains, facilitating the morphing effect across domains. Existing I2I approaches are limited to either intra-domain or deterministic inter-domain continuous translation. In this work, we present an effective signed attribute vector, which enables continuous translation on diverse mapping paths across various domains. In particular, utilizing the sign operation to encode the domain information, we introduce a unified attribute space shared by all domains, thereby allowing the interpolation on attribute vectors of different domains. To enhance the visual quality of continuous translation results, we generate a trajectory between two sign-symmetrical attribute vectors and leverage the domain information of the interpolated results along the trajectory for adversarial training. We evaluate the proposed method on a wide range of I2I translation tasks. Both qualitative and quantitative results demonstrate that the proposed framework generates more high-quality continuous translation results against the state-of-the-art methods.

2.Pushing the Envelope of Rotation Averaging for Visual SLAM ⬇️

As an essential part of structure from motion (SfM) and Simultaneous Localization and Mapping (SLAM) systems, motion averaging has been extensively studied in the past years and continues to attract surging research attention. While canonical approaches such as bundle adjustment are predominantly inherited in most of state-of-the-art SLAM systems to estimate and update the trajectory in the robot navigation, the practical implementation of bundle adjustment in SLAM systems is intrinsically limited by the high computational complexity, unreliable convergence and strict requirements of ideal initializations. In this paper, we lift these limitations and propose a novel optimization backbone for visual SLAM systems, where we leverage rotation averaging to improve the accuracy, efficiency and robustness of conventional monocular SLAM pipelines. In our approach, we first decouple the rotational and translational parameters in the camera rigid body transformation and convert the high-dimensional non-convex nonlinear problem into tractable linear subproblems in lower dimensions, and show that the subproblems can be solved independently with proper constraints. We apply the scale parameter with $l_1$-norm in the pose-graph optimization to address the rotation averaging robustness against outliers. We further validate the global optimality of our proposed approach, revisit and address the initialization schemes, pure rotational scene handling and outlier treatments. We demonstrate that our approach can exhibit up to 10x faster speed with comparable accuracy against the state of the art on public benchmarks.

3.Reducing the Annotation Effort for Video Object Segmentation Datasets ⬇️

For further progress in video object segmentation (VOS), larger, more diverse, and more challenging datasets will be necessary. However, densely labeling every frame with pixel masks does not scale to large datasets. We use a deep convolutional network to automatically create pseudo-labels on a pixel level from much cheaper bounding box annotations and investigate how far such pseudo-labels can carry us for training state-of-the-art VOS approaches. A very encouraging result of our study is that adding a manually annotated mask in only a single video frame for each object is sufficient to generate pseudo-labels which can be used to train a VOS method to reach almost the same performance level as when training with fully segmented videos. We use this workflow to create pixel pseudo-labels for the training set of the challenging tracking dataset TAO, and we manually annotate a subset of the validation set. Together, we obtain the new TAO-VOS benchmark, which we make publicly available at this http URL. While the performance of state-of-the-art methods on existing datasets starts to saturate, TAO-VOS remains very challenging for current algorithms and reveals their shortcomings.

4.SLAM in the Field: An Evaluation of Monocular Mapping and Localization on Challenging Dynamic Agricultural Environment ⬇️

This paper demonstrates a system capable of combining a sparse, indirect, monocular visual SLAM, with both offline and real-time Multi-View Stereo (MVS) reconstruction algorithms. This combination overcomes many obstacles encountered by autonomous vehicles or robots employed in agricultural environments, such as overly repetitive patterns, need for very detailed reconstructions, and abrupt movements caused by uneven roads. Furthermore, the use of a monocular SLAM makes our system much easier to integrate with an existing device, as we do not rely on a LiDAR (which is expensive and power consuming), or stereo camera (whose calibration is sensitive to external perturbation e.g. camera being displaced). To the best of our knowledge, this paper presents the first evaluation results for monocular SLAM, and our work further explores unsupervised depth estimation on this specific application scenario by simulating RGB-D SLAM to tackle the scale ambiguity, and shows our approach produces reconstructions that are helpful to various agricultural tasks. Moreover, we highlight that our experiments provide meaningful insight to improve monocular SLAM systems under agricultural settings.

5.Facial Keypoint Sequence Generation from Audio ⬇️

Whenever we speak, our voice is accompanied by facial movements and expressions. Several recent works have shown the synthesis of highly photo-realistic videos of talking faces, but they either require a source video to drive the target face or only generate videos with a fixed head pose. This lack of facial movement is because most of these works focus on the lip movement in sync with the audio while assuming the remaining facial keypoints' fixed nature. To address this, a unique audio-keypoint dataset of over 150,000 videos at 224p and 25fps is introduced that relates the facial keypoint movement for the given audio. This dataset is then further used to train the model, Audio2Keypoint, a novel approach for synthesizing facial keypoint movement to go with the audio. Given a single image of the target person and an audio sequence (in any language), Audio2Keypoint generates a plausible keypoint movement sequence in sync with the input audio, conditioned on the input image to preserve the target person's facial characteristics. To the best of our knowledge, this is the first work that proposes an audio-keypoint dataset and learns a model to output the plausible keypoint sequence to go with audio of any arbitrary length. Audio2Keypoint generalizes across unseen people with a different facial structure allowing us to generate the sequence with the voice from any source or even synthetic voices. Instead of learning a direct mapping from audio to video domain, this work aims to learn the audio-keypoint mapping that allows for in-plane and out-of-plane head rotations, while preserving the person's identity using a Pose Invariant (PIV) Encoder.

6.Multi-Task Learning for Calorie Prediction on a Novel Large-Scale Recipe Dataset Enriched with Nutritional Information ⬇️

A rapidly growing amount of content posted online, such as food recipes, opens doors to new exciting applications at the intersection of vision and language. In this work, we aim to estimate the calorie amount of a meal directly from an image by learning from recipes people have published on the Internet, thus skipping time-consuming manual data annotation. Since there are few large-scale publicly available datasets captured in unconstrained environments, we propose the pic2kcal benchmark comprising 308,000 images from over 70,000 recipes including photographs, ingredients and instructions. To obtain nutritional information of the ingredients and automatically determine the ground-truth calorie value, we match the items in the recipes with structured information from a food item database.
We evaluate various neural networks for regression of the calorie quantity and extend them with the multi-task paradigm. Our learning procedure combines the calorie estimation with prediction of proteins, carbohydrates, and fat amounts as well as a multi-label ingredient classification. Our experiments demonstrate clear benefits of multi-task learning for calorie estimation, surpassing the single-task calorie regression by 9.9%. To encourage further research on this task, we make the code for generating the dataset and the models publicly available.

7.Image Inpainting with Learnable Feature Imputation ⬇️

A regular convolution layer applying a filter in the same way over known and unknown areas causes visual artifacts in the inpainted image. Several studies address this issue with feature re-normalization on the output of the convolution. However, these models use a significant amount of learnable parameters for feature re-normalization, or assume a binary representation of the certainty of an output. We propose (layer-wise) feature imputation of the missing input values to a convolution. In contrast to learned feature re-normalization, our method is efficient and introduces a minimal number of parameters. Furthermore, we propose a revised gradient penalty for image inpainting, and a novel GAN architecture trained exclusively on adversarial loss. Our quantitative evaluation on the FDF dataset reflects that our revised gradient penalty and alternative convolution improves generated image quality significantly. We present comparisons on CelebA-HQ and Places2 to current state-of-the-art to validate our model.

8.CABiNet: Efficient Context Aggregation Network for Low-Latency Semantic Segmentation ⬇️

With the increasing demand of autonomous machines, pixel-wise semantic segmentation for visual scene understanding needs to be not only accurate but also efficient for any potential real-time applications. In this paper, we propose CABiNet (Context Aggregated Bi-lateral Network), a dual branch convolutional neural network (CNN), with significantly lower computational costs as compared to the state-of-the-art, while maintaining a competitive prediction accuracy. Building upon the existing multi-branch architectures for high-speed semantic segmentation, we design a cheap high resolution branch for effective spatial detailing and a context branch with light-weight versions of global aggregation and local distribution blocks, potent to capture both long-range and local contextual dependencies required for accurate semantic segmentation, with low computational overheads. Specifically, we achieve 76.6% and 75.9% mIOU on Cityscapes validation and test sets respectively, at 76 FPS on an NVIDIA RTX 2080Ti and 8 FPS on a Jetson Xavier NX. Codes and training models will be made publicly available.

9.PBP-Net: Point Projection and Back-Projection Network for 3D Point Cloud Segmentation ⬇️

Following considerable development in 3D scanning technologies, many studies have recently been proposed with various approaches for 3D vision tasks, including some methods that utilize 2D convolutional neural networks (CNNs). However, even though 2D CNNs have achieved high performance in many 2D vision tasks, existing works have not effectively applied them onto 3D vision tasks. In particular, segmentation has not been well studied because of the difficulty of dense prediction for each point, which requires rich feature representation. In this paper, we propose a simple and efficient architecture named point projection and back-projection network (PBP-Net), which leverages 2D CNNs for the 3D point cloud segmentation. 3 modules are introduced, each of which projects 3D point cloud onto 2D planes, extracts features using a 2D CNN backbone, and back-projects features onto the original 3D point cloud. To demonstrate effective 3D feature extraction using 2D CNN, we perform various experiments including comparison to recent methods. We analyze the proposed modules through ablation studies and perform experiments on object part segmentation (ShapeNet-Part dataset) and indoor scene semantic segmentation (S3DIS dataset). The experimental results show that proposed PBP-Net achieves comparable performance to existing state-of-the-art methods.

10.3D Multi-bodies: Fitting Sets of Plausible 3D Human Models to Ambiguous Image Data ⬇️

We consider the problem of obtaining dense 3D reconstructions of humans from single and partially occluded views. In such cases, the visual evidence is usually insufficient to identify a 3D reconstruction uniquely, so we aim at recovering several plausible reconstructions compatible with the input data. We suggest that ambiguities can be modelled more effectively by parametrizing the possible body shapes and poses via a suitable 3D model, such as SMPL for humans. We propose to learn a multi-hypothesis neural network regressor using a best-of-M loss, where each of the M hypotheses is constrained to lie on a manifold of plausible human poses by means of a generative model. We show that our method outperforms alternative approaches in ambiguous pose recovery on standard benchmarks for 3D humans, and in heavily occluded versions of these benchmarks.

11.Refactoring Policy for Compositional Generalizability using Self-Supervised Object Proposals ⬇️

We study how to learn a policy with compositional generalizability. We propose a two-stage framework, which refactorizes a high-reward teacher policy into a generalizable student policy with strong inductive bias. Particularly, we implement an object-centric GNN-based student policy, whose input objects are learned from images through self-supervised learning. Empirically, we evaluate our approach on four difficult tasks that require compositional generalizability, and achieve superior performance compared to baselines.

12.Diverse Image Captioning with Context-Object Split Latent Spaces ⬇️

Diverse image captioning models aim to learn one-to-many mappings that are innate to cross-domain datasets, such as of images and texts. Current methods for this task are based on generative latent variable models, e.g. VAEs with structured latent spaces. Yet, the amount of multimodality captured by prior work is limited to that of the paired training data -- the true diversity of the underlying generative process is not fully captured. To address this limitation, we leverage the contextual descriptions in the dataset that explain similar contexts in different visual scenes. To this end, we introduce a novel factorization of the latent space, termed context-object split, to model diversity in contextual descriptions across images and texts within the dataset. Our framework not only enables diverse captioning through context-based pseudo supervision, but extends this to images with novel objects and without paired captions in the training data. We evaluate our COS-CVAE approach on the standard COCO dataset and on the held-out COCO dataset consisting of images with novel objects, showing significant gains in accuracy and diversity.

13.Learning a Deep Reinforcement Learning Policy Over the Latent Space of a Pre-trained GAN for Semantic Age Manipulation ⬇️

Learning a disentangled representation of the latent space has become one of the most fundamental problems studied in computer vision. Recently, many generative adversarial networks (GANs) have shown promising results in generating high fidelity images. However, studies to understand the semantic layout of the latent space of pre-trained models are still limited. Several works train conditional GANs to generate faces with required semantic attributes. Unfortunately, in these attempts often the generated output is not as photo-realistic as the state of the art models. Besides, they also require large computational resources and specific datasets to generate high fidelity images. In our work, we have formulated a Markov Decision Process (MDP) over the rich latent space of a pre-trained GAN model to learn a conditional policy for semantic manipulation along specific attributes under defined identity bounds. Further, we have defined a semantic age manipulation scheme using a locally linear approximation over the latent space. Results show that our learned policy can sample high fidelity images with required age variations, while at the same time preserve the identity of the person.

14.Point Transformer ⬇️

In this work, we present Point Transformer, a deep neural network that operates directly on unordered and unstructured point sets. We design Point Transformer to extract local and global features and relate both representations by introducing the local-global attention mechanism, which aims to capture spatial point relations and shape information. For that purpose, we propose SortNet, as part of the Point Transformer, which induces input permutation invariance by selecting points based on a learned score. The output of Point Transformer is a sorted and permutation invariant feature list that can directly be incorporated into common computer vision applications. We evaluate our approach on standard classification and part segmentation benchmarks to demonstrate competitive results compared to the prior work.

15.Boost Image Captioning with Knowledge Reasoning ⬇️

Automatically generating a human-like description for a given image is a potential research in artificial intelligence, which has attracted a great of attention recently. Most of the existing attention methods explore the mapping relationships between words in sentence and regions in image, such unpredictable matching manner sometimes causes inharmonious alignments that may reduce the quality of generated captions. In this paper, we make our efforts to reason about more accurate and meaningful captions. We first propose word attention to improve the correctness of visual attention when generating sequential descriptions word-by-word. The special word attention emphasizes on word importance when focusing on different regions of the input image, and makes full use of the internal annotation knowledge to assist the calculation of visual attention. Then, in order to reveal those incomprehensible intentions that cannot be expressed straightforwardly by machines, we introduce a new strategy to inject external knowledge extracted from knowledge graph into the encoder-decoder framework to facilitate meaningful captioning. Finally, we validate our model on two freely available captioning benchmarks: Microsoft COCO dataset and Flickr30k dataset. The results demonstrate that our approach achieves state-of-the-art performance and outperforms many of the existing approaches.

16.MARNet: Multi-Abstraction Refinement Network for 3D Point Cloud Analysis ⬇️

Representation learning from 3D point clouds is challenging due to their inherent nature of permutation invariance and irregular distribution in space. Existing deep learning methods follow a hierarchical feature extraction paradigm in which high-level abstract features are derived from low-level features. However, they fail to exploit different granularity of information due to the limited interaction between these features. To this end, we propose Multi-Abstraction Refinement Network (MARNet) that ensures an effective exchange of information between multi-level features to gain local and global contextual cues while effectively preserving them till the final layer. We empirically show the effectiveness of MARNet in terms of state-of-the-art results on two challenging tasks: Shape classification and Coarse-to-fine grained semantic segmentation. MARNet significantly improves the classification performance by 2% over the baseline and outperforms the state-of-the-art methods on semantic segmentation task.

17.Facial UV Map Completion for Pose-invariant Face Recognition: A Novel Adversarial Approach based on Coupled Attention Residual UNets ⬇️

Pose-invariant face recognition refers to the problem of identifying or verifying a person by analyzing face images captured from different poses. This problem is challenging due to the large variation of pose, illumination and facial expression. A promising approach to deal with pose variation is to fulfill incomplete UV maps extracted from in-the-wild faces, then attach the completed UV map to a fitted 3D mesh and finally generate different 2D faces of arbitrary poses. The synthesized faces increase the pose variation for training deep face recognition models and reduce the pose discrepancy during the testing phase. In this paper, we propose a novel generative model called Attention ResCUNet-GAN to improve the UV map completion. We enhance the original UV-GAN by using a couple of U-Nets. Particularly, the skip connections within each U-Net are boosted by attention gates. Meanwhile, the features from two U-Nets are fused with trainable scalar weights. The experiments on the popular benchmarks, including Multi-PIE, LFW, CPLWF and CFP datasets, show that the proposed method yields superior performance compared to other existing methods.

18.Efficient texture mapping via a non-iterative global texture alignment ⬇️

Texture reconstruction techniques generally suffer from the errors in keyframe poses. We present a non-iterative method for seamless texture reconstruction of a given 3D scene. Our method finds the best texture alignment in a single shot using a global optimisation framework. First, we automatically select the best keyframe to texture each face of the mesh. This leads to a decomposition of the mesh into small groups of connected faces associated to a same keyframe. We call such groups fragments. Then, we propose a geometry-aware matching technique between the 3D keypoints extracted around the fragment borders, where the matching zone is controlled by the margin size. These constraints lead to a least squares (LS) model for finding the optimal alignment. Finally, visual seams are further reduced by applying a fast colour correction. In contrast to pixel-wise methods, we find the optimal alignment by solving a sparse system of linear equations, which is very fast and non-iterative. Experimental results demonstrate low computational complexity and outperformance compared to other alignment methods.

19.Receptive Field Size Optimization with Continuous Time Pooling ⬇️

The pooling operation is a cornerstone element of convolutional neural networks. These elements generate receptive fields for neurons, in which local perturbations should have minimal effect on the output activations, increasing robustness and invariance of the network. In this paper we will present an altered version of the most commonly applied method, maximum pooling, where pooling in theory is substituted by a continuous time differential equation, which generates a location sensitive pooling operation, more similar to biological receptive fields. We will present how this continuous method can be approximated numerically using discrete operations which fit ideally on a GPU. In our approach the kernel size is substituted by diffusion strength which is a continuous valued parameter, this way it can be optimized by gradient descent algorithms. We will evaluate the effect of continuous pooling on accuracy and computational need using commonly applied network architectures and datasets.

20.Do 2D GANs Know 3D Shape? Unsupervised 3D shape reconstruction from 2D Image GANs ⬇️

Natural images are projections of 3D objects on a 2D image plane. While state-of-the-art 2D generative models like GANs show unprecedented quality in modeling the natural image manifold, it is unclear whether they implicitly capture the underlying 3D object structures. And if so, how could we exploit such knowledge to recover the 3D shapes of objects in the images? To answer these questions, in this work, we present the first attempt to directly mine 3D geometric clues from an off-the-shelf 2D GAN that is trained on RGB images only. Through our investigation, we found that such a pre-trained GAN indeed contains rich 3D knowledge and thus can be used to recover 3D shape from a single 2D image in an unsupervised manner. The core of our framework is an iterative strategy that explores and exploits diverse viewpoint and lighting variations in the GAN image manifold. The framework does not require 2D keypoint or 3D annotations, or strong assumptions on object shapes (e.g. shapes are symmetric), yet it successfully recovers 3D shapes with high precision for human faces, cats, cars, and buildings. The recovered 3D shapes immediately allow high-quality image editing like relighting and object rotation. We quantitatively demonstrate the effectiveness of our approach compared to previous methods in both 3D shape reconstruction and face rotation. Our code and models will be released at this https URL.

21.Predicting Brain Degeneration with a Multimodal Siamese Neural Network ⬇️

To study neurodegenerative diseases, longitudinal studies are carried on volunteer patients. During a time span of several months to several years, they go through regular medical visits to acquire data from different modalities, such as biological samples, cognitive tests, structural and functional imaging. These variables are heterogeneous but they all depend on the patient's health condition, meaning that there are possibly unknown relationships between all modalities. Some information may be specific to some modalities, others may be complementary, and others may be redundant. Some data may also be missing. In this work we present a neural network architecture for multimodal learning, able to use imaging and clinical data from two time points to predict the evolution of a neurodegenerative disease, and robust to missing values. Our multimodal network achieves 92.5% accuracy and an AUC score of 0.978 over a test set of 57 subjects. We also show the superiority of the multimodal architecture, for up to 37.5% of missing values in test set subjects' clinical measurements, compared to a model using only the clinical modality.

22.PV-NAS: Practical Neural Architecture Search for Video Recognition ⬇️

Recently, deep learning has been utilized to solve video recognition problem due to its prominent representation ability. Deep neural networks for video tasks is highly customized and the design of such networks requires domain experts and costly trial and error tests. Recent advance in network architecture search has boosted the image recognition performance in a large margin. However, automatic designing of video recognition network is less explored. In this study, we propose a practical solution, namely Practical Video Neural Architecture Search (PV-NAS).Our PV-NAS can efficiently search across tremendous large scale of architectures in a novel spatial-temporal network search space using the gradient based search methods. To avoid sticking into sub-optimal solutions, we propose a novel learning rate scheduler to encourage sufficient network diversity of the searched models. Extensive empirical evaluations show that the proposed PV-NAS achieves state-of-the-art performance with much fewer computational resources. 1) Within light-weight models, our PV-NAS-L achieves 78.7% and 62.5% Top-1 accuracy on Kinetics-400 and Something-Something V2, which are better than previous state-of-the-art methods (i.e., TSM) with a large margin (4.6% and 3.4% on each dataset, respectively), and 2) among median-weight models, our PV-NAS-M achieves the best performance (also a new record)in the Something-Something V2 dataset.

23.Data-free Knowledge Distillation for Segmentation using Data-Enriching GAN ⬇️

Distilling knowledge from huge pre-trained networks to improve the performance of tiny networks has favored deep learning models to be used in many real-time and mobile applications. Several approaches that demonstrate success in this field have made use of the true training dataset to extract relevant knowledge. In absence of the True dataset, however, extracting knowledge from deep networks is still a challenge. Recent works on data-free knowledge distillation demonstrate such techniques on classification tasks. To this end, we explore the task of data-free knowledge distillation for segmentation tasks. First, we identify several challenges specific to segmentation. We make use of the DeGAN training framework to propose a novel loss function for enforcing diversity in a setting where a few classes are underrepresented. Further, we explore a new training framework for performing knowledge distillation in a data-free setting. We get an improvement of 6.93% in Mean IoU over previous approaches.

24.CaCL: Class-aware Codebook Learning for Weakly Supervised Segmentation on Diffuse Image Patterns ⬇️

Weakly supervised learning has been rapidly advanced in biomedical image analysis to achieve pixel-wise labels (segmentation) from image-wise annotations (classification), as biomedical images naturally contain image-wise labels in many scenarios. The current weakly supervised learning algorithms from the computer vision community are largely designed for focal objects (e.g., dogs and cats). However, such algorithms are not optimized for diffuse patterns in biomedical imaging (e.g., stains and fluorescent in microscopy imaging). In this paper, we propose a novel class-aware codebook learning (CaCL) algorithm to perform weakly supervised learning for diffuse image patterns. Specifically, the CaCL algorithm is deployed to segment protein expressed brush border regions from histological images of human duodenum. This paper makes the following contributions: (1) we approach the weakly supervised segmentation from a novel codebook learning perspective; (2) the CaCL algorithm segments diffuse image patterns rather than focal objects; and (3) The proposed algorithm is implemented in a multi-task framework based on Vector Quantised-Variational AutoEncoder (VQ-VAE) to perform image reconstruction, classification, feature embedding, and segmentation. The experimental results show that our method achieved superior performance compared with baseline weakly supervised algorithms.

25.A topological approach to exploring convolutional neural networks ⬇️

Motivated by the elusive understanding concerning convolution neural networks (CNNs) in view of topology, we present two theoretical frameworks to interpret two topics by using topological data analysis. The first one reveals the topological essence of CNN filters. Our theory first abstracts a topological representation of how the features locate for a CNN filter, named feature topology, and characterises it by defining the starting edge density. We reveal a principle of CNN filters: tending to organize the feature topologies for the same category, and thus propose the SED Distribution to statistically describe such an organization. We demonstrate the effectiveness of CNN filters reflects in the compactness of SED Distribution, and introduce filter entropy to measure it. Remarkably, the variation of filter entropy during training reveals the essence of CNN training: a filter-entropy-decrease process. Also, based on the principle, we give a metric to assess the filter performance. The second one investigates the inter-class distinguishability in a model-agnostic way. For each class, we propose the MBC Distribution, a distribution that could differentiate categories by characterising the intrinsic organization of the given category. As for multi-classes, we introduce the category distance which metricizes the distance between two categories, and moreover propose the CD Matrix that comprehensively evaluates not just the distinguishability between each two category pair but the distinguishable degree for each category. Finally, our experiment results confirm our theories.

26.Deep Representation Decomposition for Feature Disentanglement ⬇️

Representation disentanglement aims at learning interpretable features, so that the output can be recovered or manipulated accordingly. While existing works like infoGAN and AC-GAN exist, they choose to derive disjoint attribute code for feature disentanglement, which is not applicable for existing/trained generative models. In this paper, we propose a decomposition-GAN (dec-GAN), which is able to achieve the decomposition of an existing latent representation into content and attribute features. Guided by the classifier pre-trained on the attributes of interest, our dec-GAN decomposes the attributes of interest from the latent representation, while data recovery and feature consistency objectives enforce the learning of our proposed method. Our experiments on multiple image datasets confirm the effectiveness and robustness of our dec-GAN over recent representation disentanglement models.

27.Actor and Action Modular Network for Text-based Video Segmentation ⬇️

The actor and action semantic segmentation is a challenging problem that requires joint actor and action understanding, and learns to segment from pre-defined actor and action label pairs. However, existing methods for this task fail to distinguish those actors that have same super-category and identify the actor-action pairs that outside of the fixed actor and action vocabulary. Recent studies have extended this task using textual queries, instead of word-level actor-action pairs, to make the actor and action can be flexibly specified. In this paper, we focus on the text-based actor and action segmentation problem, which performs fine-grained actor and action understanding in the video. Previous works predicted segmentation masks from the merged heterogenous features of a given video and textual query, while they ignored that the linguistic variation of the textual query and visual semantic discrepancy of the video, and led to the asymmetric matching between convolved volumes of the video and the global query representation. To alleviate aforementioned problem, we propose a novel actor and action modular network that individually localizes the actor and action in two separate modules. We first learn the actor-/action-related content for the video and textual query, and then match them in a symmetrical manner to localize the target region. The target region includes the desired actor and action which is then fed into a fully convolutional network to predict the segmentation mask. The whole model enables joint learning for the actor-action matching and segmentation, and achieves the state-of-the-art performance on A2D Sentences and J-HMDB Sentences datasets.

28.Context-based Image Segment Labeling (CBISL) ⬇️

Working with images, one often faces problems with incomplete or unclear information. Image inpainting can be used to restore missing image regions but focuses, however, on low-level image features such as pixel intensity, pixel gradient orientation, and color. This paper aims to recover semantic image features (objects and positions) in images. Based on published gated PixelCNNs, we demonstrate a new approach referred to as quadro-directional PixelCNN to recover missing objects and return probable positions for objects based on the context. We call this approach context-based image segment labeling (CBISL). The results suggest that our four-directional model outperforms one-directional models (gated PixelCNN) and returns a human-comparable performance.

29.Set Augmented Triplet Loss for Video Person Re-Identification ⬇️

Modern video person re-identification (re-ID) machines are often trained using a metric learning approach, supervised by a triplet loss. The triplet loss used in video re-ID is usually based on so-called clip features, each aggregated from a few frame features. In this paper, we propose to model the video clip as a set and instead study the distance between sets in the corresponding triplet loss. In contrast to the distance between clip representations, the distance between clip sets considers the pair-wise similarity of each element (i.e., frame representation) between two sets. This allows the network to directly optimize the feature representation at a frame level. Apart from the commonly-used set distance metrics (e.g., ordinary distance and Hausdorff distance), we further propose a hybrid distance metric, tailored for the set-aware triplet loss. Also, we propose a hard positive set construction strategy using the learned class prototypes in a batch. Our proposed method achieves state-of-the-art results across several standard benchmarks, demonstrating the advantages of the proposed method.

30.Deep Feature Augmentation for Occluded Image Classification ⬇️

Due to the difficulty in acquiring massive task-specific occluded images, the classification of occluded images with deep convolutional neural networks (CNNs) remains highly challenging. To alleviate the dependency on large-scale occluded image datasets, we propose a novel approach to improve the classification accuracy of occluded images by fine-tuning the pre-trained models with a set of augmented deep feature vectors (DFVs). The set of augmented DFVs is composed of original DFVs and pseudo-DFVs. The pseudo-DFVs are generated by randomly adding difference vectors (DVs), extracted from a small set of clean and occluded image pairs, to the real DFVs. In the fine-tuning, the back-propagation is conducted on the DFV data flow to update the network parameters. The experiments on various datasets and network structures show that the deep feature augmentation significantly improves the classification accuracy of occluded images without a noticeable influence on the performance of clean images. Specifically, on the ILSVRC2012 dataset with synthetic occluded images, the proposed approach achieves 11.21% and 9.14% average increases in classification accuracy for the ResNet50 networks fine-tuned on the occlusion-exclusive and occlusion-inclusive training sets, respectively.

31.Mutual Information-based Disentangled Neural Networks for Classifying Unseen Categories in Different Domains: Application to Fetal Ultrasound Imaging ⬇️

Deep neural networks exhibit limited generalizability across images with different entangled domain features and categorical features. Learning generalizable features that can form universal categorical decision boundaries across domains is an interesting and difficult challenge. This problem occurs frequently in medical imaging applications when attempts are made to deploy and improve deep learning models across different image acquisition devices, across acquisition parameters or if some classes are unavailable in new training databases. To address this problem, we propose Mutual Information-based Disentangled Neural Networks (MIDNet), which extract generalizable categorical features to transfer knowledge to unseen categories in a target domain. The proposed MIDNet adopts a semi-supervised learning paradigm to alleviate the dependency on labeled data. This is important for real-world applications where data annotation is time-consuming, costly and requires training and expertise. We extensively evaluate the proposed method on fetal ultrasound datasets for two different image classification tasks where domain features are respectively defined by shadow artifacts and image acquisition devices. Experimental results show that the proposed method outperforms the state-of-the-art on the classification of unseen categories in a target domain with sparsely labeled training data.

32.CNN-Driven Quasiconformal Model for Large Deformation Image Registration ⬇️

We present a novel way to perform image registration, which is not limited to a specific kind, between image pairs with very large deformation, while preserving Quasiconformal property without tedious manual landmark labeling that conventional mathematical registration methods require. Alongside the practical function of our algorithm, one just-as-important underlying message is that the integration between typical CNN and existing Mathematical model is successful as will be pointed out by our paper, meaning that machine learning and mathematical model could coexist, cover for each other and significantly improve registration result. This paper will demonstrate an unprecedented idea of making use of both robustness of CNNs and rigorousness of mathematical model to obtain meaningful registration maps between 2D images under the aforementioned strict constraints for the sake of well-posedness.

33.Road Damage Detection using Deep Ensemble Learning ⬇️

Road damage detection is critical for the maintenance of a road, which traditionally has been performed using expensive high-performance sensors. With the recent advances in technology, especially in computer vision, it is now possible to detect and categorize different types of road damages, which can facilitate efficient maintenance and resource management. In this work, we present an ensemble model for efficient detection and classification of road damages, which we have submitted to the IEEE BigData Cup Challenge 2020. Our solution utilizes a state-of-the-art object detector known as You Only Look Once (YOLO-v4), which is trained on images of various types of road damages from Czech, Japan and India. Our ensemble approach was extensively tested with several different model versions and it was able to achieve an F1 score of 0.628 on the test 1 dataset and 0.6358 on the test 2 dataset.

34.Multi-Modal Active Learning for Automatic Liver Fibrosis Diagnosis based on Ultrasound Shear Wave Elastography ⬇️

With the development of radiomics, noninvasive diagnosis like ultrasound (US) imaging plays a very important role in automatic liver fibrosis diagnosis (ALFD). Due to the noisy data, expensive annotations of US images, the application of Artificial Intelligence (AI) assisting approaches encounters a bottleneck. Besides, the use of mono-modal US data limits the further improve of the classification results. In this work, we innovatively propose a multi-modal fusion network with active learning (MMFN-AL) for ALFD to exploit the information of multiple modalities, eliminate the noisy data and reduce the annotation cost. Four image modalities including US and three types of shear wave elastography (SWEs) are exploited. A new dataset containing these modalities from 214 candidates is well-collected and pre-processed, with the labels obtained from the liver biopsy results. Experimental results show that our proposed method outperforms the state-of-the-art performance using less than 30% data, and by using only around 80% data, the proposed fusion network achieves high AUC 89.27% and accuracy 70.59%.

35.Highway Driving Dataset for Semantic Video Segmentation ⬇️

Scene understanding is an essential technique in semantic segmentation. Although there exist several datasets that can be used for semantic segmentation, they are mainly focused on semantic image segmentation with large deep neural networks. Therefore, these networks are not useful for real time applications, especially in autonomous driving systems. In order to solve this problem, we make two contributions to semantic segmentation task. The first contribution is that we introduce the semantic video dataset, the Highway Driving dataset, which is a densely annotated benchmark for a semantic video segmentation task. The Highway Driving dataset consists of 20 video sequences having a 30Hz frame rate, and every frame is densely annotated. Secondly, we propose a baseline algorithm that utilizes a temporal correlation. Together with our attempt to analyze the temporal correlation, we expect the Highway Driving dataset to encourage research on semantic video segmentation.

36.Multi-View Adaptive Fusion Network for 3D Object Detection ⬇️

3D object detection based on LiDAR-camera fusion is becoming an emerging research theme for autonomous driving. However, it has been surprisingly difficult to effectively fuse both modalities without information loss and interference. To solve this issue, we propose a single-stage multi-view fusion framework that takes LiDAR Birds-Eye View, LiDAR Range View and Camera View images as inputs for 3D object detection. To effectively fuse multi-view features, we propose an Attentive Pointwise Fusion (APF) module to estimate the importance of the three sources with attention mechanisms which can achieve adaptive fusion of multi-view features in a pointwise manner. Besides, an Attentive Pointwise Weighting (APW) module is designed to help the network learn structure information and point feature importance with two extra tasks: foreground classification and center regression, and the predicted foreground probability will be used to reweight the point features. We design an end-to-end learnable network named MVAF-Net to integrate these two components. Our evaluations conducted on the KITTI 3D object detection datasets demonstrate that the proposed APF and APW module offer significant performance gain and that the proposed MVAF-Net achieves state-of-the-art performance in the KITTI benchmark.

37.Unsupervised Metric Relocalization Using Transform Consistency Loss ⬇️

Training networks to perform metric relocalization traditionally requires accurate image correspondences. In practice, these are obtained by restricting domain coverage, employing additional sensors, or capturing large multi-view datasets. We instead propose a self-supervised solution, which exploits a key insight: localizing a query image within a map should yield the same absolute pose, regardless of the reference image used for registration. Guided by this intuition, we derive a novel transform consistency loss. Using this loss function, we train a deep neural network to infer dense feature and saliency maps to perform robust metric relocalization in dynamic environments. We evaluate our framework on synthetic and real-world data, showing our approach outperforms other supervised methods when a limited amount of ground-truth information is available.

38.COOT: Cooperative Hierarchical Transformer for Video-Text Representation Learning ⬇️

Many real-world video-text tasks involve different levels of granularity, such as frames and words, clip and sentences or videos and paragraphs, each with distinct semantics. In this paper, we propose a Cooperative hierarchical Transformer (COOT) to leverage this hierarchy information and model the interactions between different levels of granularity and different modalities. The method consists of three major components: an attention-aware feature aggregation layer, which leverages the local temporal context (intra-level, e.g., within a clip), a contextual transformer to learn the interactions between low-level and high-level semantics (inter-level, e.g. clip-video, sentence-paragraph), and a cross-modal cycle-consistency loss to connect video and text. The resulting method compares favorably to the state of the art on several benchmarks while having few parameters. All code is available open-source at this https URL

39.FusiformNet: Extracting Discriminative Facial Features on Different Levels ⬇️

Over the last several years, research on facial recognition based on Deep Neural Network has evolved with approaches like task-specific loss functions, image normalization and augmentation, network architectures, etc. However, there have been few approaches with attention to how human faces differ from person to person. Premising that inter-personal differences are found both generally and locally on the human face, I propose FusiformNet, a novel framework for feature extraction that leverages the nature of person-identifying facial features. Tested on ImageUnrestricted setting of Labeled Face in the Wild benchmark, this method achieved a state-of-the-art accuracy of 96.67% without labeled outside data, image augmentation, normalization, or special loss functions. Likewise, the method also performed on par with previous state-of-the-arts when pretrained on CASIA-WebFace dataset. Considering its ability to extract both general and local facial features, the utility of FusiformNet may not be limited to facial recognition but also extend to other DNN-based tasks.

40.Human Leg Motion Tracking by Fusing IMUs and RGB Camera Data Using Extended Kalman Filter ⬇️

Human motion capture is frequently used to study rehabilitation and clinical problems, as well as to provide realistic animation for the entertainment industry. IMU-based systems, as well as Marker-based motion tracking systems, are the most popular methods to track movement due to their low cost of implementation and lightweight. This paper proposes a quaternion-based Extended Kalman filter approach to recover the human leg segments motions with a set of IMU sensors data fused with camera-marker system data. In this paper, an Extended Kalman Filter approach is developed to fuse the data of two IMUs and one RGB camera for human leg motion tracking. Based on the complementary properties of the inertial sensors and camera-marker system, in the introduced new measurement model, the orientation data of the upper leg and the lower leg is updated through three measurement equations. The positioning of the human body is made possible by the tracked position of the pelvis joint by the camera marker system. A mathematical model has been utilized to estimate joints' depth in 2D images. The efficiency of the proposed algorithm is evaluated by an optical motion tracker system.

41.DeepOpht: Medical Report Generation for Retinal Images via Deep Models and Visual Explanation ⬇️

In this work, we propose an AI-based method that intends to improve the conventional retinal disease treatment procedure and help ophthalmologists increase diagnosis efficiency and accuracy. The proposed method is composed of a deep neural networks-based (DNN-based) module, including a retinal disease identifier and clinical description generator, and a DNN visual explanation module. To train and validate the effectiveness of our DNN-based module, we propose a large-scale retinal disease image dataset. Also, as ground truth, we provide a retinal image dataset manually labeled by ophthalmologists to qualitatively show, the proposed AI-based method is effective. With our experimental results, we show that the proposed method is quantitatively and qualitatively effective. Our method is capable of creating meaningful retinal image descriptions and visual explanations that are clinically relevant.

42.LG-GAN: Label Guided Adversarial Network for Flexible Targeted Attack of Point Cloud-based Deep Networks ⬇️

Deep neural networks have made tremendous progress in 3D point-cloud recognition. Recent works have shown that these 3D recognition networks are also vulnerable to adversarial samples produced from various attack methods, including optimization-based 3D Carlini-Wagner attack, gradient-based iterative fast gradient method, and skeleton-detach based point-dropping. However, after a careful analysis, these methods are either extremely slow because of the optimization/iterative scheme, or not flexible to support targeted attack of a specific category. To overcome these shortcomings, this paper proposes a novel label guided adversarial network (LG-GAN) for real-time flexible targeted point cloud attack. To the best of our knowledge, this is the first generation based 3D point cloud attack method. By feeding the original point clouds and target attack label into LG-GAN, it can learn how to deform the point clouds to mislead the recognition network into the specific label only with a single forward pass. In detail, LGGAN first leverages one multi-branch adversarial network to extract hierarchical features of the input point clouds, then incorporates the specified label information into multiple intermediate features using the label encoder. Finally, the encoded features will be fed into the coordinate reconstruction decoder to generate the target adversarial sample. By evaluating different point-cloud recognition models (e.g., PointNet, PointNet++ and DGCNN), we demonstrate that the proposed LG-GAN can support flexible targeted attack on the fly while guaranteeing good attack performance and higher efficiency simultaneously.

43.Memory Group Sampling Based Online Action Recognition Using Kinetic Skeleton Features ⬇️

Online action recognition is an important task for human centered intelligent services, which is still difficult to achieve due to the varieties and uncertainties of spatial and temporal scales of human actions. In this paper, we propose two core ideas to handle the online action recognition problem. First, we combine the spatial and temporal skeleton features to depict the actions, which include not only the geometrical features, but also multi-scale motion features, such that both the spatial and temporal information of the action are covered. Second, we propose a memory group sampling method to combine the previous action frames and current action frames, which is based on the truth that the neighbouring frames are largely redundant, and the sampling mechanism ensures that the long-term contextual information is also considered. Finally, an improved 1D CNN network is employed for training and testing using the features from sampled frames. The comparison results to the state of the art methods using the public datasets show that the proposed method is fast and efficient, and has competitive performance

44.Adversarial Self-Supervised Scene Flow Estimation ⬇️

This work proposes a metric learning approach for self-supervised scene flow estimation. Scene flow estimation is the task of estimating 3D flow vectors for consecutive 3D point clouds. Such flow vectors are fruitful, \eg for recognizing actions, or avoiding collisions. Training a neural network via supervised learning for scene flow is impractical, as this requires manual annotations for each 3D point at each new timestamp for each scene. To that end, we seek for a self-supervised approach, where a network learns a latent metric to distinguish between points translated by flow estimations and the target point cloud. Our adversarial metric learning includes a multi-scale triplet loss on sequences of two-point clouds as well as a cycle consistency loss. Furthermore, we outline a benchmark for self-supervised scene flow estimation: the Scene Flow Sandbox. The benchmark consists of five datasets designed to study individual aspects of flow estimation in progressive order of complexity, from a moving object to real-world scenes. Experimental evaluation on the benchmark shows that our approach obtains state-of-the-art self-supervised scene flow results, outperforming recent neighbor-based approaches. We use our proposed benchmark to expose shortcomings and draw insights on various training setups. We find that our setup captures motion coherence and preserves local geometries. Dealing with occlusions, on the other hand, is still an open challenge.

45.Autonomous Extraction of Gleason Patterns for Grading Prostate Cancer using Multi-Gigapixel Whole Slide Images ⬇️

Prostate cancer (PCa) is the second deadliest form of cancer in males. The severity of PCa can be clinically graded through the Gleason scores obtained by examining the structural representation of Gleason cellular patterns. This paper presents an asymmetric encoder-decoder model that integrates a novel hierarchical decomposition block to exploit the feature representations pooled across various scales and then fuses them together to generate the Gleason cellular patterns using the whole slide images. Furthermore, the proposed network is penalized through a novel three-tiered hybrid loss function which ensures that the proposed model accurately recognizes the cluttered regions of the cancerous tissues despite having similar contextual and textural characteristics. We have rigorously tested the proposed network on 10,516 whole slide scans (containing around 71.7M patches), where the proposed model achieved 3.59% improvement over state-of-the-art scene parsing, encoder-decoder, and fully convolutional networks in terms of intersection-over-union.

46.HM4: Hidden Markov Model with Memory Management for Visual Place Recognition ⬇️

Visual place recognition needs to be robust against appearance variability due to natural and man-made causes. Training data collection should thus be an ongoing process to allow continuous appearance changes to be recorded. However, this creates an unboundedly-growing database that poses time and memory scalability challenges for place recognition methods. To tackle the scalability issue for visual place recognition in autonomous driving, we develop a Hidden Markov Model approach with a two-tiered memory management. Our algorithm, dubbed HM$^4$, exploits temporal look-ahead to transfer promising candidate images between passive storage and active memory when needed. The inference process takes into account both promising images and a coarse representations of the full database. We show that this allows constant time and space inference for a fixed coverage area. The coarse representations can also be updated incrementally to absorb new data. To further reduce the memory requirements, we derive a compact image representation inspired by Locality Sensitive Hashing (LSH). Through experiments on real world data, we demonstrate the excellent scalability and accuracy of the approach under appearance changes and provide comparisons against state-of-the-art techniques.

47.A Parallel Approach for Real-Time Face Recognition from a Large Database ⬇️

We present a new facial recognition system, capable of identifying a person, provided their likeness has been previously stored in the system, in real time. The system is based on storing and comparing facial embeddings of the subject, and identifying them later within a live video feed. This system is highly accurate, and is able to tag people with their ID in real time. It is able to do so, even when using a database containing thousands of facial embeddings, by using a parallelized searching technique. This makes the system quite fast and allows it to be highly scalable.

48.Efficient Pipelines for Vision-Based Context Sensing ⬇️

Context awareness is an essential part of mobile and ubiquitous computing. Its goal is to unveil situational information about mobile users like locations and activities. The sensed context can enable many services like navigation, AR, and smarting shopping. Such context can be sensed in different ways including visual sensors. There is an emergence of vision sources deployed worldwide. The cameras could be installed on roadside, in-house, and on mobile platforms. This trend provides huge amount of vision data that could be used for context sensing. However, the vision data collection and analytics are still highly manual today. It is hard to deploy cameras at large scale for data collection. Organizing and labeling context from the data are also labor intensive. In recent years, advanced vision algorithms and deep neural networks are used to help analyze vision data. But this approach is limited by data quality, labeling effort, and dependency on hardware resources. In summary, there are three major challenges for today's vision-based context sensing systems: data collection and labeling at large scale, process large data volumes efficiently with limited hardware resources, and extract accurate context out of vision data. The thesis explores the design space that consists of three dimensions: sensing task, sensor types, and task locations. Our prior work explores several points in this design space. We make contributions by (1) developing efficient and scalable solutions for different points in the design space of vision-based sensing tasks; (2) achieving state-of-the-art accuracy in those applications; (3) and developing guidelines for designing such sensing systems.

49.Dark Reciprocal-Rank: Boosting Graph-Convolutional Self-Localization Network via Teacher-to-student Knowledge Transfer ⬇️

In visual robot self-localization, graph-based scene representation and matching have recently attracted research interest as robust and discriminative methods for selflocalization. Although effective, their computational and storage costs do not scale well to large-size environments. To alleviate this problem, we formulate self-localization as a graph classification problem and attempt to use the graph convolutional neural network (GCN) as a graph classification engine. A straightforward approach is to use visual feature descriptors that are employed by state-of-the-art self-localization systems, directly as graph node features. However, their superior performance in the original self-localization system may not necessarily be replicated in GCN-based self-localization. To address this issue, we introduce a novel teacher-to-student knowledge-transfer scheme based on rank matching, in which the reciprocal-rank vector output by an off-the-shelf state-of-the-art teacher self-localization model is used as the dark knowledge to transfer. Experiments indicate that the proposed graph-convolutional self-localization network can significantly outperform state-of-the-art self-localization systems, as well as the teacher classifier.

50.Temporally-Continuous Probabilistic Prediction using Polynomial Trajectory Parameterization ⬇️

A commonly-used representation for motion prediction of actors is a sequence of waypoints (comprising positions and orientations) for each actor at discrete future time-points. While this approach is simple and flexible, it can exhibit unrealistic higher-order derivatives (such as acceleration) and approximation errors at intermediate time steps. To address this issue we propose a simple and general representation for temporally continuous probabilistic trajectory prediction that is based on polynomial trajectory parameterization. We evaluate the proposed representation on supervised trajectory prediction tasks using two large self-driving data sets. The results show realistic higher-order derivatives and better accuracy at interpolated time-points, as well as the benefits of the inferred noise distributions over the trajectories. Extensive experimental studies based on existing state-of-the-art models demonstrate the effectiveness of the proposed approach relative to other representations in predicting the future motions of vehicle, bicyclist, and pedestrian traffic actors.

51.IndRNN Based Long-term Temporal Recognition in the Spatial and Frequency Domain ⬇️

Smartphone sensors based human activity recognition is attracting increasing interests nowadays with the popularization of smartphones. With the high sampling rates of smartphone sensors, it is a highly long-range temporal recognition problem, especially with the large intra-class distances such as the smartphones carried at different locations such as in the bag or on the body, and the small inter-class distances such as taking train or subway. To address this problem, we propose a new approach, an Independently Recurrent Neural Network (IndRNN) based long-term temporal activity recognition with spatial and frequency domain features. Considering the periodic characteristics of the sensor data, short-term temporal features are first extracted in the spatial and frequency domains. Then the IndRNN, which is able to capture long-term patterns, is used to further obtain the long-term features for classification. In view of the large differences when the smartphone is carried at different locations, a group based location recognition is first developed to pinpoint the location of the smartphone. The Sussex-Huawei Locomotion (SHL) dataset from the SHL Challenge is used for evaluation. An earlier version of the proposed method has won the second place award in the SHL Challenge 2020 (the first place if not considering multiple models fusion approach). The proposed method is further improved in this paper and achieves 80.72$%$ accuracy, better than the existing methods using a single model.

52.Real-Time Text Detection and Recognition ⬇️

Inrecentyears,ConvolutionalNeuralNet-work(CNN) is quite a popular topic, as it is a powerful andintelligent technique that can be applied in various fields.The YOLO is a technique that uses the algorithms for real-time text detection tasks. However, issues like, photometricdistortion and geometric distortion, could affect the systemYOLO accuracy and cause system failure. Therefore, thereare improvements that can make the system work better. Inthis paper, we are going to present our solution - a potentialsolution of a fast and accurate real-time text direction andrecognition system. The paper covers the topic of Real-TimeText detection and recognition in three major areas: 1. videoand image preprocess, 2. Text detection, 3. Text recognition. Asa mature technique, there are many existing methods that canpotentially improve the solution. We will go through some ofthose existing methods in the literature review session. In thisway, we are presenting an industrial strength, high-accuracy,Real-Time Text Detection and recognition tool.

53.A Survey on Contrastive Self-supervised Learning ⬇️

Self-supervised learning has gained popularity because of its ability to avoid the cost of annotating large-scale datasets. It is capable of adopting self-defined pseudo labels as supervision and use the learned representations for several downstream tasks. Specifically, contrastive learning has recently become a dominant component in self-supervised learning methods for computer vision, natural language processing (NLP), and other domains. It aims at embedding augmented versions of the same sample close to each other while trying to push away embeddings from different samples. This paper provides an extensive review of self-supervised methods that follow the contrastive approach. The work explains commonly used pretext tasks in a contrastive learning setup, followed by different architectures that have been proposed so far. Next, we have a performance comparison of different methods for multiple downstream tasks such as image classification, object detection, and action recognition. Finally, we conclude with the limitations of the current methods and the need for further techniques and future directions to make substantial progress.

54.TartanVO: A Generalizable Learning-based VO ⬇️

We present the first learning-based visual odometry (VO) model, which generalizes to multiple datasets and real-world scenarios and outperforms geometry-based methods in challenging scenes. We achieve this by leveraging the SLAM dataset TartanAir, which provides a large amount of diverse synthetic data in challenging environments. Furthermore, to make our VO model generalize across datasets, we propose an up-to-scale loss function and incorporate the camera intrinsic parameters into the model. Experiments show that a single model, TartanVO, trained only on synthetic data, without any finetuning, can be generalized to real-world datasets such as KITTI and EuRoC, demonstrating significant advantages over the geometry-based methods on challenging trajectories. Our code is available at this https URL.

55.Unsupervised Deep Persistent Monocular Visual Odometry and Depth Estimation in Extreme Environments ⬇️

In recent years, unsupervised deep learning approaches have received significant attention to estimate the depth and visual odometry (VO) from unlabelled monocular image sequences. However, their performance is limited in challenging environments due to perceptual degradation, occlusions and rapid motions. Moreover, the existing unsupervised methods suffer from the lack of scale-consistency constraints across frames, which causes that the VO estimators fail to provide persistent trajectories over long sequences. In this study, we propose an unsupervised monocular deep VO framework that predicts six-degrees-of-freedom pose camera motion and depth map of the scene from unlabelled RGB image sequences. We provide detailed quantitative and qualitative evaluations of the proposed framework on a) a challenging dataset collected during the DARPA Subterranean challenge; and b) the benchmark KITTI and Cityscapes datasets. The proposed approach outperforms both traditional and state-of-the-art unsupervised deep VO methods providing better results for both pose estimation and depth recovery. The presented approach is part of the solution used by the COSTAR team participating at the DARPA Subterranean Challenge.

56.Self-paced and self-consistent co-training for semi-supervised image segmentation ⬇️

Deep co-training has recently been proposed as an effective approach for image segmentation when annotated data is scarce. In this paper, we improve existing approaches for semi-supervised segmentation with a self-paced and self-consistent co-training method. To help distillate information from unlabeled images, we first design a self-paced learning strategy for co-training that lets jointly-trained neural networks focus on easier-to-segment regions first, and then gradually consider harder ones.This is achieved via an end-to-end differentiable loss inthe form of a generalized Jensen Shannon Divergence(JSD). Moreover, to encourage predictions from different networks to be both consistent and confident, we enhance this generalized JSD loss with an uncertainty regularizer based on entropy. The robustness of individual models is further improved using a self-ensembling loss that enforces their prediction to be consistent across different training iterations. We demonstrate the potential of our method on three challenging image segmentation problems with different image modalities, using small fraction of labeled data. Results show clear advantages in terms of performance compared to the standard co-training baselines and recently proposed state-of-the-art approaches for semi-supervised segmentation

57.Scene Flow from Point Clouds with or without Learning ⬇️

Scene flow is the three-dimensional (3D) motion field of a scene. It provides information about the spatial arrangement and rate of change of objects in dynamic environments. Current learning-based approaches seek to estimate the scene flow directly from point clouds and have achieved state-of-the-art performance. However, supervised learning methods are inherently domain specific and require a large amount of labeled data. Annotation of scene flow on real-world point clouds is expensive and challenging, and the lack of such datasets has recently sparked interest in self-supervised learning methods. How to accurately and robustly learn scene flow representations without labeled real-world data is still an open problem. Here we present a simple and interpretable objective function to recover the scene flow from point clouds. We use the graph Laplacian of a point cloud to regularize the scene flow to be "as-rigid-as-possible". Our proposed objective function can be used with or without learning---as a self-supervisory signal to learn scene flow representations, or as a non-learning-based method in which the scene flow is optimized during runtime. Our approach outperforms related works in many datasets. We also show the immediate applications of our proposed method for two applications: motion segmentation and point cloud densification.

58.General Data Analytics With Applications To Visual Information Analysis: A Provable Backward-Compatible Semisimple Paradigm Over T-Algebra ⬇️

We consider a novel backward-compatible paradigm of general data analytics over a recently-reported semisimple algebra (called t-algebra). We study the abstract algebraic framework over the t-algebra by representing the elements of t-algebra by fix-sized multi-way arrays of complex numbers and the algebraic structure over the t-algebra by a collection of direct-product constituents. Over the t-algebra, many algorithms, if not all, are generalized in a straightforward manner using this new semisimple paradigm. To demonstrate the new paradigm's performance and its backward-compatibility, we generalize some canonical algorithms for visual pattern analysis. Experiments on public datasets show that the generalized algorithms compare favorably with their canonical counterparts.

59.Pose Randomization for Weakly Paired Image Style Translation ⬇️

Utilizing the trained model under different conditions without data annotation is attractive for robot applications. Towards this goal, one class of methods is to translate the image style from the training environment to the current one. Conventional studies on image style translation mainly focus on two settings: paired data on images from two domains with exactly aligned content, and unpaired data, with independent content. In this paper, we would like to propose a new setting, where the content in the two images is aligned with error in poses. We consider that this setting is more practical since robots with various sensors are able to align the data up to some error level, even with different styles. To solve this problem, we propose PRoGAN to learn a style translator by intentionally transforming the original domain images with a noisy pose, then matching the distribution of translated transformed images and the distribution of the target domain images. The adversarial training enforces the network to learn the style translation, avoiding being entangled with other variations. In addition, we propose two pose estimation based self-supervised tasks to further improve the performance. Finally, PRoGAN is validated on both simulated and real-world collected data to show the effectiveness. Results on down-stream tasks, classification, road segmentation, object detection, and feature matching show its potential for real applications. this https URL .

60.LandmarkGAN: Synthesizing Faces from Landmarks ⬇️

Face synthesis is an important problem in computer vision with many applications. In this work, we describe a new method, namely LandmarkGAN, to synthesize faces based on facial landmarks as input. Facial landmarks are a natural, intuitive, and effective representation for facial expressions and orientations, which are independent from the target's texture or color and background scene. Our method is able to transform a set of facial landmarks into new faces of different subjects, while retains the same facial expression and orientation. Experimental results on face synthesis and reenactments demonstrate the effectiveness of our method.

61.ProxylessKD: Direct Knowledge Distillation with Inherited Classifier for Face Recognition ⬇️

Knowledge Distillation (KD) refers to transferring knowledge from a large model to a smaller one, which is widely used to enhance model performance in machine learning. It tries to align embedding spaces generated from the teacher and the student model (i.e. to make images corresponding to the same semantics share the same embedding across different models). In this work, we focus on its application in face recognition. We observe that existing knowledge distillation models optimize the proxy tasks that force the student to mimic the teacher's behavior, instead of directly optimizing the face recognition accuracy. Consequently, the obtained student models are not guaranteed to be optimal on the target task or able to benefit from advanced constraints, such as large margin constraints (e.g. margin-based softmax). We then propose a novel method named ProxylessKD that directly optimizes face recognition accuracy by inheriting the teacher's classifier as the student's classifier to guide the student to learn discriminative embeddings in the teacher's embedding space. The proposed ProxylessKD is very easy to implement and sufficiently generic to be extended to other tasks beyond face recognition. We conduct extensive experiments on standard face recognition benchmarks, and the results demonstrate that ProxylessKD achieves superior performance over existing knowledge distillation methods.

62.Temporal Smoothing for 3D Human Pose Estimation and Localization for Occluded People ⬇️

In multi-person pose estimation actors can be heavily occluded, even become fully invisible behind another person. While temporal methods can still predict a reasonable estimation for a temporarily disappeared pose using past and future frames, they exhibit large errors nevertheless. We present an energy minimization approach to generate smooth, valid trajectories in time, bridging gaps in visibility. We show that it is better than other interpolation based approaches and achieves state of the art results. In addition, we present the synthetic MuCo-Temp dataset, a temporal extension of the MuCo-3DHP dataset. Our code is made publicly available.

63.Enhanced Balancing GAN: Minority-class Image Generation ⬇️

Generative adversarial networks (GANs) are one of the most powerful generative models, but always require a large and balanced dataset to train. Traditional GANs are not applicable to generate minority-class images in a highly imbalanced dataset. Balancing GAN (BAGAN) is proposed to mitigate this problem, but it is unstable when images in different classes look similar, e.g. flowers and cells. In this work, we propose a supervised autoencoder with an intermediate embedding model to disperse the labeled latent vectors. With the improved autoencoder initialization, we also build an architecture of BAGAN with gradient penalty (BAGAN-GP). Our proposed model overcomes the unstable issue in original BAGAN and converges faster to high quality generations. Our model achieves high performance on the imbalanced scale-down version of MNIST Fashion, CIFAR-10, and one small-scale medical image dataset.

64.Self-supervised Representation Learning for Evolutionary Neural Architecture Search ⬇️

Recently proposed neural architecture search (NAS) algorithms adopt neural predictors to accelerate the architecture search. The capability of neural predictors to accurately predict the performance metrics of neural architecture is critical to NAS, and the acquisition of training datasets for neural predictors is time-consuming. How to obtain a neural predictor with high prediction accuracy using a small amount of training data is a central problem to neural predictor-based NAS. Here, we firstly design a new architecture encoding scheme that overcomes the drawbacks of existing vector-based architecture encoding schemes to calculate the graph edit distance of neural architectures. To enhance the predictive performance of neural predictors, we devise two self-supervised learning methods from different perspectives to pre-train the architecture embedding part of neural predictors to generate a meaningful representation of neural architectures. The first one is to train a carefully designed two branch graph neural network model to predict the graph edit distance of two input neural architectures. The second method is inspired by the prevalently contrastive learning, and we present a new contrastive learning algorithm that utilizes a central feature vector as a proxy to contrast positive pairs against negative pairs. Experimental results illustrate that the pre-trained neural predictors can achieve comparable or superior performance compared with their supervised counterparts with several times less training samples. We achieve state-of-the-art performance on the NASBench-101 and NASBench201 benchmarks when integrating the pre-trained neural predictors with an evolutionary NAS algorithm.

65.Exploring Severe Occlusion: Multi-Person 3D Pose Estimation with Gated Convolution ⬇️

3D human pose estimation (HPE) is crucial in many fields, such as human behavior analysis, augmented reality/virtual reality (AR/VR) applications, and self-driving industry. Videos that contain multiple potentially occluded people captured from freely moving monocular cameras are very common in real-world scenarios, while 3D HPE for such scenarios is quite challenging, partially because there is a lack of such data with accurate 3D ground truth labels in existing datasets. In this paper, we propose a temporal regression network with a gated convolution module to transform 2D joints to 3D and recover the missing occluded joints in the meantime. A simple yet effective localization approach is further conducted to transform the normalized pose to the global trajectory. To verify the effectiveness of our approach, we also collect a new moving camera multi-human (MMHuman) dataset that includes multiple people with heavy occlusion captured by moving cameras. The 3D ground truth joints are provided by accurate motion capture (MoCap) system. From the experiments on static-camera based Human3.6M data and our own collected moving-camera based data, we show that our proposed method outperforms most state-of-the-art 2D-to-3D pose estimation methods, especially for the scenarios with heavy occlusions.

66.Learning Open Set Network with Discriminative Reciprocal Points ⬇️

Open set recognition is an emerging research area that aims to simultaneously classify samples from predefined classes and identify the rest as 'unknown'. In this process, one of the key challenges is to reduce the risk of generalizing the inherent characteristics of numerous unknown samples learned from a small amount of known data. In this paper, we propose a new concept, Reciprocal Point, which is the potential representation of the extra-class space corresponding to each known category. The sample can be classified to known or unknown by the otherness with reciprocal points. To tackle the open set problem, we offer a novel open space risk regularization term. Based on the bounded space constructed by reciprocal points, the risk of unknown is reduced through multi-category interaction. The novel learning framework called Reciprocal Point Learning (RPL), which can indirectly introduce the unknown information into the learner with only known classes, so as to learn more compact and discriminative representations. Moreover, we further construct a new large-scale challenging aircraft dataset for open set recognition: Aircraft 300 (Air-300). Extensive experiments on multiple benchmark datasets indicate that our framework is significantly superior to other existing approaches and achieves state-of-the-art performance on standard open set benchmarks.

67.Multimodal and self-supervised representation learning for automatic gesture recognition in surgical robotics ⬇️

Self-supervised, multi-modal learning has been successful in holistic representation of complex scenarios. This can be useful to consolidate information from multiple modalities which have multiple, versatile uses. Its application in surgical robotics can lead to simultaneously developing a generalised machine understanding of the surgical process and reduce the dependency on quality, expert annotations which are generally difficult to obtain. We develop a self-supervised, multi-modal representation learning paradigm that learns representations for surgical gestures from video and kinematics. We use an encoder-decoder network configuration that encodes representations from surgical videos and decodes them to yield kinematics. We quantitatively demonstrate the efficacy of our learnt representations for gesture recognition (with accuracy between 69.6 % and 77.8 %), transfer learning across multiple tasks (with accuracy between 44.6 % and 64.8 %) and surgeon skill classification (with accuracy between 76.8 % and 81.2 %). Further, we qualitatively demonstrate that our self-supervised representations cluster in semantically meaningful properties (surgeon skill and gestures).

68.Automatic Chronic Degenerative Diseases Identification Using Enteric Nervous System Images ⬇️

Studies recently accomplished on the Enteric Nervous System have shown that chronic degenerative diseases affect the Enteric Glial Cells (EGC) and, thus, the development of recognition methods able to identify whether or not the EGC are affected by these type of diseases may be helpful in its diagnoses. In this work, we propose the use of pattern recognition and machine learning techniques to evaluate if a given animal EGC image was obtained from a healthy individual or one affect by a chronic degenerative disease. In the proposed approach, we have performed the classification task with handcrafted features and deep learning based techniques, also known as non-handcrafted features. The handcrafted features were obtained from the textural content of the ECG images using texture descriptors, such as the Local Binary Pattern (LBP). Moreover, the representation learning techniques employed in the approach are based on different Convolutional Neural Network (CNN) architectures, such as AlexNet and VGG16, with and without transfer learning. The complementarity between the handcrafted and non-handcrafted features was also evaluated with late fusion techniques. The datasets of EGC images used in the experiments, which are also contributions of this paper, are composed of three different chronic degenerative diseases: Cancer, Diabetes Mellitus, and Rheumatoid Arthritis. The experimental results, supported by statistical analysis, shown that the proposed approach can distinguish healthy cells from the sick ones with a recognition rate of 89.30% (Rheumatoid Arthritis), 98.45% (Cancer), and 95.13% (Diabetes Mellitus), being achieved by combining classifiers obtained both feature scenarios.

69.Weakly Supervised 3D Classification of Chest CT using Aggregated Multi-Resolution Deep Segmentation Features ⬇️

Weakly supervised disease classification of CT imaging suffers from poor localization owing to case-level annotations, where even a positive scan can hold hundreds to thousands of negative slices along multiple planes. Furthermore, although deep learning segmentation and classification models extract distinctly unique combinations of anatomical features from the same target class(es), they are typically seen as two independent processes in a computer-aided diagnosis (CAD) pipeline, with little to no feature reuse. In this research, we propose a medical classifier that leverages the semantic structural concepts learned via multi-resolution segmentation feature maps, to guide weakly supervised 3D classification of chest CT volumes. Additionally, a comparative analysis is drawn across two different types of feature aggregation to explore the vast possibilities surrounding feature fusion. Using a dataset of 1593 scans labeled on a case-level basis via rule-based model, we train a dual-stage convolutional neural network (CNN) to perform organ segmentation and binary classification of four representative diseases (emphysema, pneumonia/atelectasis, mass and nodules) in lungs. The baseline model, with separate stages for segmentation and classification, results in AUC of 0.791. Using identical hyperparameters, the connected architecture using static and dynamic feature aggregation improves performance to AUC of 0.832 and 0.851, respectively. This study advances the field in two key ways. First, case-level report data is used to weakly supervise a 3D CT classifier of multiple, simultaneous diseases for an organ. Second, segmentation and classification models are connected with two different feature aggregation strategies to enhance the classification performance.

70.Leveraging Adaptive Color Augmentation in Convolutional Neural Networks for Deep Skin Lesion Segmentation ⬇️

Fully automatic detection of skin lesions in dermatoscopic images can facilitate early diagnosis and repression of malignant melanoma and non-melanoma skin cancer. Although convolutional neural networks are a powerful solution, they are limited by the illumination spectrum of annotated dermatoscopic screening images, where color is an important discriminative feature. In this paper, we propose an adaptive color augmentation technique to amplify data expression and model performance, while regulating color difference and saturation to minimize the risks of using synthetic data. Through deep visualization, we qualitatively identify and verify the semantic structural features learned by the network for discriminating skin lesions against normal skin tissue. The overall system achieves a Dice Ratio of 0.891 with 0.943 sensitivity and 0.932 specificity on the ISIC 2018 Testing Set for segmentation.

71.Pixel-Level Cycle Association: A New Perspective for Domain Adaptive Semantic Segmentation ⬇️

Domain adaptive semantic segmentation aims to train a model performing satisfactory pixel-level predictions on the target with only out-of-domain (source) annotations. The conventional solution to this task is to minimize the discrepancy between source and target to enable effective knowledge transfer. Previous domain discrepancy minimization methods are mainly based on the adversarial training. They tend to consider the domain discrepancy globally, which ignore the pixel-wise relationships and are less discriminative. In this paper, we propose to build the pixel-level cycle association between source and target pixel pairs and contrastively strengthen their connections to diminish the domain gap and make the features more discriminative. To the best of our knowledge, this is a new perspective for tackling such a challenging task. Experiment results on two representative domain adaptation benchmarks, i.e. GTAV $\rightarrow$ Cityscapes and SYNTHIA $\rightarrow$ Cityscapes, verify the effectiveness of our proposed method and demonstrate that our method performs favorably against previous state-of-the-arts. Our method can be trained end-to-end in one stage and introduces no additional parameters, which is expected to serve as a general framework and help ease future research in domain adaptive semantic segmentation. Code is available at this https URL Level-Cycle-Association.

72.(Un)Masked COVID-19 Trends from Social Media ⬇️

COVID-19 has affected the entire world. One useful protection method for people against COVID-19 is to wear masks in public areas. Across the globe, many public service providers have mandated correctly wearing masks to use their services. This paper proposes two new datasets VAriety MAsks - Classification VAMA-C) and VAriety MAsks - Segmentation (VAMA-S), for mask detection and mask fit analysis tasks, respectively. We propose a framework for classifying masked and unmasked faces and a segmentation based model to calculate the mask-fit score. Both the models trained in this study achieved an accuracy of 98%. Using the two trained deep learning models, 2.04 million social media images for six major US cities were analyzed. By comparing the regulations, an increase in masks worn in images as the COVID-19 cases rose in these cities was observed, particularly when their respective states imposed strict regulations. Furthermore, mask compliance in the Black Lives Matter protest was analyzed, eliciting that 40% of the people in group photos wore masks, and 45% of them wore the masks with a fit score of greater than 80%.

73.Pose-based Body Language Recognition for Emotion and Psychiatric Symptom Interpretation ⬇️

Inspired by the human ability to infer emotions from body language, we propose an automated framework for body language based emotion recognition starting from regular RGB videos. In collaboration with psychologists, we further extend the framework for psychiatric symptom prediction. Because a specific application domain of the proposed framework may only supply a limited amount of data, the framework is designed to work on a small training set and possess a good transferability. The proposed system in the first stage generates sequences of body language predictions based on human poses estimated from input videos. In the second stage, the predicted sequences are fed into a temporal network for emotion interpretation and psychiatric symptom prediction. We first validate the accuracy and transferability of the proposed body language recognition method on several public action recognition datasets. We then evaluate the framework on a proposed URMC dataset, which consists of conversations between a standardized patient and a behavioral health professional, along with expert annotations of body language, emotions, and potential psychiatric symptoms. The proposed framework outperforms other methods on the URMC dataset.

74.A Deep Learning Study on Osteosarcoma Detection from Histological Images ⬇️

In the U.S, 5-10% of new pediatric cases of cancer are primary bone tumors. The most common type of primary malignant bone tumor is osteosarcoma. The intention of the present work is to improve the detection and diagnosis of osteosarcoma using computer-aided detection (CAD) and diagnosis (CADx). Such tools as convolutional neural networks (CNNs) can significantly decrease the surgeon's workload and make a better prognosis of patient conditions. CNNs need to be trained on a large amount of data in order to achieve a more trustworthy performance. In this study, transfer learning techniques, pre-trained CNNs, are adapted to a public dataset on osteosarcoma histological images to detect necrotic images from non-necrotic and healthy tissues. First, the dataset was preprocessed, and different classifications are applied. Then, Transfer learning models including VGG19 and Inception V3 are used and trained on Whole Slide Images (WSI) with no patches, to improve the accuracy of the outputs. Finally, the models are applied to different classification problems, including binary and multi-class classifiers. Experimental results show that the accuracy of the VGG19 has the highest, 96%, performance amongst all binary classes and multiclass classification. Our fine-tuned model demonstrates state-of-the-art performance on detecting malignancy of Osteosarcoma based on histologic images.

75.Perceive, Attend, and Drive: Learning Spatial Attention for Safe Self-Driving ⬇️

In this paper, we propose an end-to-end self-driving network featuring a sparse attention module that learns to automatically attend to important regions of the input. The attention module specifically targets motion planning, whereas prior literature only applied attention in perception tasks. Learning an attention mask directly targeted for motion planning significantly improves the planner safety by performing more focused computation. Furthermore, visualizing the attention improves interpretability of end-to-end self-driving.

76.Into the Wild with AudioScope: Unsupervised Audio-Visual Separation of On-Screen Sounds ⬇️

Recent progress in deep learning has enabled many advances in sound separation and visual scene understanding. However, extracting sound sources which are apparent in natural videos remains an open problem. In this work, we present AudioScope, a novel audio-visual sound separation framework that can be trained without supervision to isolate on-screen sound sources from real in-the-wild videos. Prior audio-visual separation work assumed artificial limitations on the domain of sound classes (e.g., to speech or music), constrained the number of sources, and required strong sound separation or visual segmentation labels. AudioScope overcomes these limitations, operating on an open domain of sounds, with variable numbers of sources, and without labels or prior visual segmentation. The training procedure for AudioScope uses mixture invariant training (MixIT) to separate synthetic mixtures of mixtures (MoMs) into individual sources, where noisy labels for mixtures are provided by an unsupervised audio-visual coincidence model. Using the noisy labels, along with attention between video and audio features, AudioScope learns to identify audio-visual similarity and to suppress off-screen sounds. We demonstrate the effectiveness of our approach using a dataset of video clips extracted from open-domain YFCC100m video data. This dataset contains a wide diversity of sound classes recorded in unconstrained conditions, making the application of previous methods unsuitable. For evaluation and semi-supervised experiments, we collected human labels for presence of on-screen and off-screen sounds on a small subset of clips.

77.U-Net and its variants for medical image segmentation: theory and applications ⬇️

U-net is an image segmentation technique developed primarily for medical image analysis that can precisely segment images using a scarce amount of training data. These traits provide U-net with a very high utility within the medical imaging community and have resulted in extensive adoption of U-net as the primary tool for segmentation tasks in medical imaging. The success of U-net is evident in its widespread use in all major image modalities from CT scans and MRI to X-rays and microscopy. Furthermore, while U-net is largely a segmentation tool, there have been instances of the use of U-net in other applications. As the potential of U-net is still increasing, in this review we look at the various developments that have been made in the U-net architecture and provide observations on recent trends. We examine the various innovations that have been made in deep learning and discuss how these tools facilitate U-net. Furthermore, we look at image modalities and application areas where U-net has been applied.

78.Top 10 BraTS 2020 challenge solution: Brain tumor segmentation with self-ensembled, deeply-supervised 3D-Unet like neural networks ⬇️

Brain tumor segmentation is a critical task for patient's disease management. To this end, we trained multiple U-net like neural networks, mainly with deep supervision and stochastic weight averaging, on the Multimodal Brain Tumor Segmentation Challenge (BraTS) 2020 training dataset, in a cross-validated fashion. Final brain tumor segmentations were produced by first averaging independently two sets of models, and then custom merging the labelmaps to account for individual performance of each set. Our performance on the online validation dataset with test time augmentation were as follows: Dice of 0.81, 0.91 and 0.85; Hausdorff (95%) of 20.6, 4,3, 5.7 mm for the enhancing tumor, whole tumor and tumor core, respectively. Similarly, our ensemble achieved a Dice of 0.79, 0.89 and 0.84, as well as Hausdorff (95%) of 20.4, 6.7 and 19.5mm on the final test dataset. More complicated training schemes and neural network architectures were investigated, without significant performance gain, at the cost of greatly increased training time. While relatively straightforward, our approach yielded good and balanced performance for each tumor subregions. Our solution is open sourced at this https URL.

79.Depth Ranging Performance Evaluation and Improvement for RGB-D Cameras on Field-Based High-Throughput Phenotyping Robots ⬇️

RGB-D cameras have been successfully used for indoor High-ThroughpuT Phenotyping (HTTP). However, their capability and feasibility for in-field HTTP still need to be evaluated, due to the noise and disturbances generated by unstable illumination, specular reflection, and diffuse reflection, etc. To solve these problems, we evaluated the depth-ranging performances of two consumer-level RGB-D cameras (RealSense D435i and Kinect V2) under in-field HTTP scenarios, and proposed a strategy to compensate the depth measurement error. For performance evaluation, we focused on determining their optimal ranging areas for different crop organs. Based on the evaluation results, we proposed a brightness-and-distance-based Support Vector Regression Strategy, to compensate the ranging error. Furthermore, we analyzed the depth filling rate of two RGB-D cameras under different lighting intensities. Experimental results showed that: 1) For RealSense D435i, its effective ranging area is [0.160, 1.400] m, and in-field filling rate is approximately 90%. 2) For Kinect V2, it has a high ranging accuracy in the [0.497, 1.200] m, but its in-field filling rate is less than 24.9%. 3) Our error compensation model can effectively reduce the influences of lighting intensity and target distance. The maximum MSE and minimum R2 of this model are 0.029 and 0.867, respectively. To sum up, RealSense D435i has better ranging performances than Kinect V2 on in-field HTTP.

80.AVECL-UMONS database for audio-visual event classification and localization ⬇️

We introduce the AVECL-UMons dataset for audio-visual event classification and localization in the context of office environments. The audio-visual dataset is composed of 11 event classes recorded at several realistic positions in two different rooms. Two types of sequences are recorded according to the number of events in the sequence. The dataset comprises 2662 unilabel sequences and 2724 multilabel sequences corresponding to a total of 5.24 hours. The dataset is publicly accessible online : this https URL.

81.ASIST: Annotation-free synthetic instance segmentation and tracking for microscope video analysis ⬇️

Instance object segmentation and tracking provide comprehensive quantification of objects across microscope videos. The recent single-stage pixel-embedding based deep learning approach has shown its superior performance compared with "segment-then-associate" two-stage solutions. However, one major limitation of applying a supervised pixel-embedding based method to microscope videos is the resource-intensive manual labeling, which involves tracing hundreds of overlapped objects with their temporal associations across video frames. Inspired by the recent generative adversarial network (GAN) based annotation-free image segmentation, we propose a novel annotation-free synthetic instance segmentation and tracking (ASIST) algorithm for analyzing microscope videos of sub-cellular microvilli. The contributions of this paper are three-fold: (1) proposing a new annotation-free video analysis paradigm is proposed. (2) aggregating the embedding based instance segmentation and tracking with annotation-free synthetic learning as a holistic framework; and (3) to the best of our knowledge, this is first study to investigate microvilli instance segmentation and tracking using embedding based deep learning. From the experimental results, the proposed annotation-free method achieved superior performance compared with supervised learning.

82.Deep Learning in Computer-Aided Diagnosis and Treatment of Tumors: A Survey ⬇️

Computer-Aided Diagnosis and Treatment of Tumors is a hot topic of deep learning in recent years, which constitutes a series of medical tasks, such as detection of tumor markers, the outline of tumor leisures, subtypes and stages of tumors, prediction of therapeutic effect, and drug development. Meanwhile, there are some deep learning models with precise positioning and excellent performance produced in mainstream task scenarios. Thus follow to introduce deep learning methods from task-orient, mainly focus on the improvements for medical tasks. Then to summarize the recent progress in four stages of tumor diagnosis and treatment, which named In-Vitro Diagnosis (IVD), Imaging Diagnosis (ID), Pathological Diagnosis (PD), and Treatment Planning (TP). According to the specific data types and medical tasks of each stage, we present the applications of deep learning in the Computer-Aided Diagnosis and Treatment of Tumors and analyzing the excellent works therein. This survey concludes by discussing research issues and suggesting challenges for future improvement.

83.nnU-Net for Brain Tumor Segmentation ⬇️

We apply nnU-Net to the segmentation task of the BraTS 2020 challenge. The unmodified nnU-Net baseline configuration already achieves a respectable result. By incorporating BraTS-specific modifications regarding postprocessing, region-based training, a more aggressive data augmentation as well as several minor modifications to the nnUNet pipeline we are able to improve its segmentation performance substantially. We furthermore re-implement the BraTS ranking scheme to determine which of our nnU-Net variants best fits the requirements imposed by it. Our final ensemble took the first place in the BraTS 2020 competition with Dice scores of 88.95, 85.06 and 82.03 and HD95 values of 8.498,17.337 and 17.805 for whole tumor, tumor core and enhancing tumor, respectively.

84.Bifurcated Autoencoder for Segmentation of COVID-19 Infected Regions in CT Images ⬇️

The new coronavirus infection has shocked the world since early 2020 with its aggressive outbreak. Rapid detection of the disease saves lives, and relying on medical imaging (Computed Tomography and X-ray) to detect infected lungs has shown to be effective. Deep learning and convolutional neural networks have been used for image analysis in this context. However, accurate identification of infected regions has proven challenging for two main reasons. Firstly, the characteristics of infected areas differ in different images. Secondly, insufficient training data makes it challenging to train various machine learning algorithms, including deep-learning models. This paper proposes an approach to segment lung regions infected by COVID-19 to help cardiologists diagnose the disease more accurately, faster, and more manageable. We propose a bifurcated 2-D model for two types of segmentation. This model uses a shared encoder and a bifurcated connection to two separate decoders. One decoder is for segmentation of the healthy region of the lungs, while the other is for the segmentation of the infected regions. Experiments on publically available images show that the bifurcated structure segments infected regions of the lungs better than state of the art.

85.Brain Tumor Classification Using Medial Residual Encoder Layers ⬇️

According to the World Health Organization, cancer is the second leading cause of death worldwide, responsible for over 9.5 million deaths in 2018 alone. Brain tumors count for one out of every four cancer deaths. Accurate and timely diagnosis of brain tumors will lead to more effective treatments. To date, several image classification approaches have been proposed to aid diagnosis and treatment. We propose an encoder layer that uses post-max-pooling features for residual learning. Our approach shows promising results by improving the tumor classification accuracy in MR images using a limited medical image dataset. Experimental evaluations of this model on a dataset consisting of 3064 MR images show 95-98% accuracy, which is better than previous studies on this database.

86.Tracking Partially-Occluded Deformable Objects while Enforcing Geometric Constraints ⬇️

In order to manipulate a deformable object, such as rope or cloth, in unstructured environments, robots need a way to estimate its current shape. However, tracking the shape of a deformable object can be challenging because of the object's high flexibility, (self-)occlusion, and interaction with obstacles. Building a high-fidelity physics simulation to aid in tracking is difficult for novel environments. Instead we focus on tracking the object based on RGBD images and geometric motion estimates and obstacles. Our key contributions over previous work in this vein are: 1) A better way to handle severe occlusion by using a motion model to regularize the tracking estimate; and 2) The formulation of \textit{convex} geometric constraints, which allow us to prevent self-intersection and penetration into known obstacles via a post-processing step. These contributions allow us to outperform previous methods by a large margin in terms of accuracy in scenarios with severe occlusion and obstacles.

87.Triage of Potential COVID-19 Patients from Chest X-ray Images using Hierarchical Convolutional Networks ⬇️

The current COVID-19 pandemic has motivated the researchers to use artificial intelligence techniques for potential alternatives to reverse transcription polymerase chain reaction (RT-PCR) due to the limited scale of testing. The chest X-ray (CXR) is one of the alternatives to achieve fast diagnosis but the unavailability of large scale annotated data makes the clinical implementation of machine learning-based COVID detection methods difficult. Another important issue is the usage of ImageNet pre-trained networks which does not guarantee to extract reliable feature representations. In this paper, we propose the use of hierarchical convolutional network (HCN) architecture to naturally augment the data along with diversified features. The HCN uses the first convolution layer from COVIDNet followed by the convolutional layers from well known pre-trained networks to extract the features. The use of the convolution layer from COVIDNet ensures the extraction of representations relevant to the CXR modality. We also propose the use of ECOC for encoding multiclass problems to binary classification for improving the recognition performance. Experimental results show that HCN architecture is capable of achieving better results in comparison to the existing studies. The proposed method can accurately triage potential COVID-19 patients through CXR images for sharing the testing load and increasing the testing capacity.

88.Learning Euler's Elastica Model for Medical Image Segmentation ⬇️

Image segmentation is a fundamental topic in image processing and has been studied for many decades. Deep learning-based supervised segmentation models have achieved state-of-the-art performance but most of them are limited by using pixel-wise loss functions for training without geometrical constraints. Inspired by Euler's Elastica model and recent active contour models introduced into the field of deep learning, we propose a novel active contour with elastica (ACE) loss function incorporating Elastica (curvature and length) and region information as geometrically-natural constraints for the image segmentation tasks. We introduce the mean curvature i.e. the average of all principal curvatures, as a more effective image prior to representing curvature in our ACE loss function. Furthermore, based on the definition of the mean curvature, we propose a fast solution to approximate the ACE loss in three-dimensional (3D) by using Laplace operators for 3D image segmentation. We evaluate our ACE loss function on four 2D and 3D natural and biomedical image datasets. Our results show that the proposed loss function outperforms other mainstream loss functions on different segmentation networks. Our source code is available at this https URL.

89.Generating Correct Answers for Progressive Matrices Intelligence Tests ⬇️

Raven's Progressive Matrices are multiple-choice intelligence tests, where one tries to complete the missing location in a $3\times 3$ grid of abstract images. Previous attempts to address this test have focused solely on selecting the right answer out of the multiple choices. In this work, we focus, instead, on generating a correct answer given the grid, without seeing the choices, which is a harder task, by definition. The proposed neural model combines multiple advances in generative models, including employing multiple pathways through the same network, using the reparameterization trick along two pathways to make their encoding compatible, a dynamic application of variational losses, and a complex perceptual loss that is coupled with a selective backpropagation procedure. Our algorithm is able not only to generate a set of plausible answers, but also to be competitive to the state of the art methods in multiple-choice tests.

90.Dynamic radiomics: a new methodology to extract quantitative time-related features from tomographic images ⬇️

The feature extraction methods of radiomics are mainly based on static tomographic images at a certain moment, while the occurrence and development of disease is a dynamic process that cannot be fully reflected by only static characteristics. This study proposes a new dynamic radiomics feature extraction workflow that uses time-dependent tomographic images of the same patient, focuses on the changes in image features over time, and then quantifies them as new dynamic features for diagnostic or prognostic evaluation. We first define the mathematical paradigm of dynamic radiomics and introduce three specific methods that can describe the transformation process of features over time. Three different clinical problems are used to validate the performance of the proposed dynamic feature with conventional 2D and 3D static features.

91.Two-layer clustering-based sparsifying transform learning for low-dose CT reconstruction ⬇️

Achieving high-quality reconstructions from low-dose computed tomography (LDCT) measurements is of much importance in clinical settings. Model-based image reconstruction methods have been proven to be effective in removing artifacts in LDCT. In this work, we propose an approach to learn a rich two-layer clustering-based sparsifying transform model (MCST2), where image patches and their subsequent feature maps (filter residuals) are clustered into groups with different learned sparsifying filters per group. We investigate a penalized weighted least squares (PWLS) approach for LDCT reconstruction incorporating learned MCST2 priors. Experimental results show the superior performance of the proposed PWLS-MCST2 approach compared to other related recent schemes.

92.Using Monte Carlo dropout and bootstrap aggregation for uncertainty estimation in radiation therapy dose prediction with deep learning neural networks ⬇️

Recently, artificial intelligence technologies and algorithms have become a major focus for advancements in treatment planning for radiation therapy. As these are starting to become incorporated into the clinical workflow, a major concern from clinicians is not whether the model is accurate, but whether the model can express to a human operator when it does not know if its answer is correct. We propose to use Monte Carlo dropout (MCDO) and the bootstrap aggregation (bagging) technique on deep learning models to produce uncertainty estimations for radiation therapy dose prediction. We show that both models are capable of generating a reasonable uncertainty map, and, with our proposed scaling technique, creating interpretable uncertainties and bounds on the prediction and any relevant metrics. Performance-wise, bagging provides statistically significant reduced loss value and errors in most of the metrics investigated in this study. The addition of bagging was able to further reduce errors by another 0.34% for Dmean and 0.19% for Dmax, on average, when compared to the baseline framework. Overall, the bagging framework provided significantly lower MAE of 2.62, as opposed to the baseline framework's MAE of 2.87. The usefulness of bagging, from solely a performance standpoint, does highly depend on the problem and the acceptable predictive error, and its high upfront computational cost during training should be factored in to deciding whether it is advantageous to use it. In terms of deployment with uncertainty estimations turned on, both frameworks offer the same performance time of about 12 seconds. As an ensemble-based metaheuristic, bagging can be used with existing machine learning architectures to improve stability and performance, and MCDO can be applied to any deep learning models that have dropout as part of their architecture.

93.Segmentation of Infrared Breast Images Using MultiResUnet Neural Network ⬇️

Breast cancer is the second leading cause of death for women in the U.S. Early detection of breast cancer is key to higher survival rates of breast cancer patients. We are investigating infrared (IR) thermography as a noninvasive adjunct to mammography for breast cancer screening. IR imaging is radiation-free, pain-free, and non-contact. Automatic segmentation of the breast area from the acquired full-size breast IR images will help limit the area for tumor search, as well as reduce the time and effort costs of manual segmentation. Autoencoder-like convolutional and deconvolutional neural networks (C-DCNN) had been applied to automatically segment the breast area in IR images in previous studies. In this study, we applied a state-of-the-art deep-learning segmentation model, MultiResUnet, which consists of an encoder part to capture features and a decoder part for precise localization. It was used to segment the breast area by using a set of breast IR images, collected in our pilot study by imaging breast cancer patients and normal volunteers with a thermal infrared camera (N2 Imager). The database we used has 450 images, acquired from 14 patients and 16 volunteers. We used a thresholding method to remove interference in the raw images and remapped them from the original 16-bit to 8-bit, and then cropped and segmented the 8-bit images manually. Experiments using leave-one-out cross-validation (LOOCV) and comparison with the ground-truth images by using Tanimoto similarity show that the average accuracy of MultiResUnet is 91.47%, which is about 2% higher than that of the autoencoder. MultiResUnet offers a better approach to segment breast IR images than our previous model.

94.Pose Estimation of Specular and Symmetrical Objects ⬇️

In the robotic industry, specular and textureless metallic components are ubiquitous. The 6D pose estimation of such objects with only a monocular RGB camera is difficult because of the absence of rich texture features. Furthermore, the appearance of specularity heavily depends on the camera viewpoint and environmental light conditions making traditional methods, like template matching, fail. In the last 30 years, pose estimation of the specular object has been a consistent challenge, and most related works require massive knowledge modeling effort for light setups, environment, or the object surface. On the other hand, recent works exhibit the feasibility of 6D pose estimation on a monocular camera with convolutional neural networks(CNNs) however they mostly use opaque objects for evaluation. This paper provides a data-driven solution to estimate the 6D pose of specular objects for grasping them, proposes a cost function for handling symmetry, and demonstrates experimental results showing the system's feasibility.

95.DL-Reg: A Deep Learning Regularization Technique using Linear Regression ⬇️

Regularization plays a vital role in the context of deep learning by preventing deep neural networks from the danger of overfitting. This paper proposes a novel deep learning regularization method named as DL-Reg, which carefully reduces the nonlinearity of deep networks to a certain extent by explicitly enforcing the network to behave as much linear as possible. The key idea is to add a linear constraint to the objective function of the deep neural networks, which is simply the error of a linear mapping from the inputs to the outputs of the model. More precisely, the proposed DL-Reg carefully forces the network to behave in a linear manner. This linear constraint, which is further adjusted by a regularization factor, prevents the network from the risk of overfitting. The performance of DL-Reg is evaluated by training state-of-the-art deep network models on several benchmark datasets. The experimental results show that the proposed regularization method: 1) gives major improvements over the existing regularization techniques, and 2) significantly improves the performance of deep neural networks, especially in the case of small-sized training datasets.

96.Deep learning in the ultrasound evaluation of neonatal respiratory status ⬇️

Lung ultrasound imaging is reaching growing interest from the scientific community. On one side, thanks to its harmlessness and high descriptive power, this kind of diagnostic imaging has been largely adopted in sensitive applications, like the diagnosis and follow-up of preterm newborns in neonatal intensive care units. On the other side, state-of-the-art image analysis and pattern recognition approaches have recently proven their ability to fully exploit the rich information contained in these data, making them attractive for the research community. In this work, we present a thorough analysis of recent deep learning networks and training strategies carried out on a vast and challenging multicenter dataset comprising 87 patients with different diseases and gestational ages. These approaches are employed to assess the lung respiratory status from ultrasound images and are evaluated against a reference marker. The conducted analysis sheds some light on this problem by showing the critical points that can mislead the training procedure and proposes some adaptations to the specific data and task. The achieved results sensibly outperform those obtained by a previous work, which is based on textural features, and narrow the gap with the visual score predicted by the human experts.

97.Encoding Clinical Priori in 3D Convolutional Neural Networks for Prostate Cancer Detection in bpMRI ⬇️

We hypothesize that anatomical priors can be viable mediums to infuse domain-specific clinical knowledge into state-of-the-art convolutional neural networks (CNN) based on the U-Net architecture. We introduce a probabilistic population prior which captures the spatial prevalence and zonal distinction of clinically significant prostate cancer (csPCa), in order to improve its computer-aided detection (CAD) in bi-parametric MR imaging (bpMRI). To evaluate performance, we train 3D adaptations of the U-Net, U-SEResNet, UNet++ and Attention U-Net using 800 institutional training-validation scans, paired with radiologically-estimated annotations and our computed prior. For 200 independent testing bpMRI scans with histologically-confirmed delineations of csPCa, our proposed method of encoding clinical priori demonstrates a strong ability to improve patient-based diagnosis (upto 8.70% increase in AUROC) and lesion-level detection (average increase of 1.08 pAUC between 0.1-1.0 false positive per patient) across all four architectures.

98.Meta-Learning with Adaptive Hyperparameters ⬇️

Despite its popularity, several recent works question the effectiveness of MAML when test tasks are different from training tasks, thus suggesting various task-conditioned methodology to improve the initialization. Instead of searching for better task-aware initialization, we focus on a complementary factor in MAML framework, inner-loop optimization (or fast adaptation). Consequently, we propose a new weight update rule that greatly enhances the fast adaptation process. Specifically, we introduce a small meta-network that can adaptively generate per-step hyperparameters: learning rate and weight decay coefficients. The experimental results validate that the Adaptive Learning of hyperparameters for Fast Adaptation (ALFA) is the equally important ingredient that was often neglected in the recent few-shot learning approaches. Surprisingly, fast adaptation from random initialization with ALFA can already outperform MAML.

99.Combining Domain-Specific Meta-Learners in the Parameter Space for Cross-Domain Few-Shot Classification ⬇️

The goal of few-shot classification is to learn a model that can classify novel classes using only a few training examples. Despite the promising results shown by existing meta-learning algorithms in solving the few-shot classification problem, there still remains an important challenge: how to generalize to unseen domains while meta-learning on multiple seen domains? In this paper, we propose an optimization-based meta-learning method, called Combining Domain-Specific Meta-Learners (CosML), that addresses the cross-domain few-shot classification problem. CosML first trains a set of meta-learners, one for each training domain, to learn prior knowledge (i.e., meta-parameters) specific to each domain. The domain-specific meta-learners are then combined in the \emph{parameter space}, by taking a weighted average of their meta-parameters, which is used as the initialization parameters of a task network that is quickly adapted to novel few-shot classification tasks in an unseen domain. Our experiments show that CosML outperforms a range of state-of-the-art methods and achieves strong cross-domain generalization ability.

100.Evaluation of Inference Attack Models for Deep Learning on Medical Data ⬇️

Deep learning has attracted broad interest in healthcare and medical communities. However, there has been little research into the privacy issues created by deep networks trained for medical applications. Recently developed inference attack algorithms indicate that images and text records can be reconstructed by malicious parties that have the ability to query deep networks. This gives rise to the concern that medical images and electronic health records containing sensitive patient information are vulnerable to these attacks. This paper aims to attract interest from researchers in the medical deep learning community to this important problem. We evaluate two prominent inference attack models, namely, attribute inference attack and model inversion attack. We show that they can reconstruct real-world medical images and clinical reports with high fidelity. We then investigate how to protect patients' privacy using defense mechanisms, such as label perturbation and model perturbation. We provide a comparison of attack results between the original and the medical deep learning models with defenses. The experimental evaluations show that our proposed defense approaches can effectively reduce the potential privacy leakage of medical deep learning from the inference attacks.

101.Dense Pixel-wise Micro-motion Estimation of Object Surface by using Low Dimensional Embedding of Laser Speckle Pattern ⬇️

This paper proposes a method of estimating micro-motion of an object at each pixel that is too small to detect under a common setup of camera and illumination. The method introduces an active-lighting approach to make the motion visually detectable. The approach is based on speckle pattern, which is produced by the mutual interference of laser light on object's surface and continuously changes its appearance according to the out-of-plane motion of the surface. In addition, speckle pattern becomes uncorrelated with large motion. To compensate such micro- and large motion, the method estimates the motion parameters up to scale at each pixel by nonlinear embedding of the speckle pattern into low-dimensional space. The out-of-plane motion is calculated by making the motion parameters spatially consistent across the image. In the experiments, the proposed method is compared with other measuring devices to prove the effectiveness of the method.

102.Integer Programming-based Error-Correcting Output Code Design for Robust Classification ⬇️

Error-Correcting Output Codes (ECOCs) offer a principled approach for combining simple binary classifiers into multiclass classifiers. In this paper, we investigate the problem of designing optimal ECOCs to achieve both nominal and adversarial accuracy using Support Vector Machines (SVMs) and binary deep learning models. In contrast to previous literature, we present an Integer Programming (IP) formulation to design minimal codebooks with desirable error correcting properties. Our work leverages the advances in IP solvers to generate codebooks with optimality guarantees. To achieve tractability, we exploit the underlying graph-theoretic structure of the constraint set in our IP formulation. This enables us to use edge clique covers to substantially reduce the constraint set. Our codebooks achieve a high nominal accuracy relative to standard codebooks (e.g., one-vs-all, one-vs-one, and dense/sparse codes). We also estimate the adversarial accuracy of our ECOC-based classifiers in a white-box setting. Our IP-generated codebooks provide non-trivial robustness to adversarial perturbations even without any adversarial training.

103.EDCNN: Edge enhancement-based Densely Connected Network with Compound Loss for Low-Dose CT Denoising ⬇️

In the past few decades, to reduce the risk of X-ray in computed tomography (CT), low-dose CT image denoising has attracted extensive attention from researchers, which has become an important research issue in the field of medical images. In recent years, with the rapid development of deep learning technology, many algorithms have emerged to apply convolutional neural networks to this task, achieving promising results. However, there are still some problems such as low denoising efficiency, over-smoothed result, etc. In this paper, we propose the Edge enhancement based Densely connected Convolutional Neural Network (EDCNN). In our network, we design an edge enhancement module using the proposed novel trainable Sobel convolution. Based on this module, we construct a model with dense connections to fuse the extracted edge information and realize end-to-end image denoising. Besides, when training the model, we introduce a compound loss that combines MSE loss and multi-scales perceptual loss to solve the over-smoothed problem and attain a marked improvement in image quality after denoising. Compared with the existing low-dose CT image denoising algorithms, our proposed model has a better performance in preserving details and suppressing noise.

104.Multi-stage transfer learning for lung segmentation using portable X-ray devices for patients with COVID-19 ⬇️

In 2020, the SARS-CoV-2 virus causes a global pandemic of the new human coronavirus disease COVID-19. This pathogen primarily infects the respiratory system of the afflicted, usually resulting in pneumonia and in a severe case of acute respiratory distress syndrome. These disease developments result in the formation of different pathological structures in the lungs, similar to those observed in other viral pneumonias that can be detected by the use of chest X-rays. For this reason, the detection and analysis of the pulmonary regions, the main focus of affection of COVID-19, becomes a crucial part of both clinical and automatic diagnosis processes. Due to the overload of the health services, portable X-ray devices are widely used, representing an alternative to fixed devices to reduce the risk of cross-contamination. However, these devices entail different complications as the image quality that, together with the subjectivity of the clinician, make the diagnostic process more difficult. In this work, we developed a novel fully automatic methodology specially designed for the identification of these lung regions in X-ray images of low quality as those from portable devices. To do so, we took advantage of a large dataset from magnetic resonance imaging of a similar pathology and performed two stages of transfer learning to obtain a robust methodology with a low number of images from portable X-ray devices. This way, our methodology obtained a satisfactory accuracy of $0.9761 \pm 0.0100$ for patients with COVID-19, $0.9801 \pm 0.0104$ for normal patients and $0.9769 \pm 0.0111$ for patients with pulmonary diseases with similar characteristics as COVID-19 (such as pneumonia) but not genuine COVID-19.

105.C-Net: A Reliable Convolutional Neural Network for Biomedical Image Classification ⬇️

Cancers are the leading cause of death in many developed countries. Early diagnosis plays a crucial role in having proper treatment for this debilitating disease. The automated classification of the type of cancer is a challenging task since pathologists must examine a huge number of histopathological images to detect infinitesimal abnormalities. In this study, we propose a novel convolutional neural network (CNN) architecture composed of a Concatenation of multiple Networks, called C-Net, to classify biomedical images. In contrast to conventional deep learning models in biomedical image classification, which utilize transfer learning to solve the problem, no prior knowledge is employed. The model incorporates multiple CNNs including Outer, Middle, and Inner. The first two parts of the architecture contain six networks that serve as feature extractors to feed into the Inner network to classify the images in terms of malignancy and benignancy. The C-Net is applied for histopathological image classification on two public datasets, including BreakHis and Osteosarcoma. To evaluate the performance, the model is tested using several evaluation metrics for its reliability. The C-Net model outperforms all other models on the individual metrics for both datasets and achieves zero misclassification.

106.83% ImageNet Accuracy in One Hour ⬇️

EfficientNets are a family of state-of-the-art image classification models based on efficiently scaled convolutional neural networks. Currently, EfficientNets can take on the order of days to train; for example, training an EfficientNet-B0 model takes 23 hours on a Cloud TPU v2-8 node. In this paper, we explore techniques to scale up the training of EfficientNets on TPU-v3 Pods with 2048 cores, motivated by speedups that can be achieved when training at such scales. We discuss optimizations required to scale training to a batch size of 65536 on 1024 TPU-v3 cores, such as selecting large batch optimizers and learning rate schedules as well as utilizing distributed evaluation and batch normalization techniques. Additionally, we present timing and performance benchmarks for EfficientNet models trained on the ImageNet dataset in order to analyze the behavior of EfficientNets at scale. With our optimizations, we are able to train EfficientNet on ImageNet to an accuracy of 83% in 1 hour and 4 minutes.

107.Adversarial Robust Training in MRI Reconstruction ⬇️

Deep Learning has shown potential in accelerating Magnetic Resonance Image acquisition and reconstruction. Nevertheless, there is a dearth of tailored methods to guarantee that the reconstruction of small features is achieved with high fidelity. In this work, we employ adversarial attacks to generate small synthetic perturbations that when added to the input MRI, they are not reconstructed by a trained DL reconstruction network. Then, we use robust training to increase the network's sensitivity to small features and encourage their reconstruction. Next, we investigate the generalization of said approach to real world features. For this, a musculoskeletal radiologist annotated a set of cartilage and meniscal lesions from the knee Fast-MRI dataset, and a classification network was devised to assess the features reconstruction. Experimental results show that by introducing robust training to a reconstruction network, the rate (4.8%) of false negative features in image reconstruction can be reduced. The results are encouraging and highlight the necessity for attention on this problem by the image reconstruction community, as a milestone for the introduction of DL reconstruction in clinical practice. To support further research, we make our annotation publicly available at this https URL.