1.Near real-time map building with multi-class image set labelling and classification of road conditions using convolutional neural networks ⬇️
Weather is an important factor affecting transportation and road safety. In this paper, we leverage state-of-the-art convolutional neural networks in labelling images taken by street and highway cameras located across across North America. Road camera snapshots were used in experiments with multiple deep learning frameworks to classify images by road condition. The training data for these experiments used images labelled as dry, wet, snow/ice, poor, and offline. The experiments tested different configurations of six convolutional neural networks (VGG-16, ResNet50, Xception, InceptionResNetV2, EfficientNet-B0 and EfficientNet-B4) to assess their suitability to this problem. The precision, accuracy, and recall were measured for each framework configuration. In addition, the training sets were varied both in overall size and by size of individual classes. The final training set included 47,000 images labelled using the five aforementioned classes. The EfficientNet-B4 framework was found to be most suitable to this problem, achieving validation accuracy of 90.6%, although EfficientNet-B0 achieved an accuracy of 90.3% with half the execution time. It was observed that VGG-16 with transfer learning proved to be very useful for data acquisition and pseudo-labelling with limited hardware resources, throughout this project. The EfficientNet-B4 framework was then placed into a real-time production environment, where images could be classified in real-time on an ongoing basis. The classified images were then used to construct a map showing real-time road conditions at various camera locations across North America. The choice of these frameworks and our analysis take into account unique requirements of real-time map building functions. A detailed analysis of the process of semi-automated dataset labelling using these frameworks is also presented in this paper.
2.Depthwise-STFT based separable Convolutional Neural Networks ⬇️
In this paper, we propose a new convolutional layer called Depthwise-STFT Separable layer that can serve as an alternative to the standard depthwise separable convolutional layer. The construction of the proposed layer is inspired by the fact that the Fourier coefficients can accurately represent important features such as edges in an image. It utilizes the Fourier coefficients computed (channelwise) in the 2D local neighborhood (e.g., 3x3) of each position of the input map to obtain the feature maps. The Fourier coefficients are computed using 2D Short Term Fourier Transform (STFT) at multiple fixed low frequency points in the 2D local neighborhood at each position. These feature maps at different frequency points are then linearly combined using trainable pointwise (1x1) convolutions. We show that the proposed layer outperforms the standard depthwise separable layer-based models on the CIFAR-10 and CIFAR-100 image classification datasets with reduced space-time complexity.
3.DRMIME: Differentiable Mutual Information and Matrix Exponential for Multi-Resolution Image Registration ⬇️
In this work, we present a novel unsupervised image registration algorithm. It is differentiable end-to-end and can be used for both multi-modal and mono-modal registration. This is done using mutual information (MI) as a metric. The novelty here is that rather than using traditional ways of approximating MI, we use a neural estimator called MINE and supplement it with matrix exponential for transformation matrix computation. This leads to improved results as compared to the standard algorithms available out-of-the-box in state-of-the-art image registration toolboxes.
4.Handling noise in image deblurring via joint learning ⬇️
Currently, many blind deblurring methods assume blurred images are noise-free and perform unsatisfactorily on the blurry images with noise. Unfortunately, noise is quite common in real scenes. A straightforward solution is to denoise images before deblurring them. However, even state-of-the-art denoisers cannot guarantee to remove noise entirely. Slight residual noise in the denoised images could cause significant artifacts in the deblurring stage. To tackle this problem, we propose a cascaded framework consisting of a denoiser subnetwork and a deblurring subnetwork. In contrast to previous methods, we train the two subnetworks jointly. Joint learning reduces the effect of the residual noise after denoising on deblurring, hence improves the robustness of deblurring to heavy noise. Moreover, our method is also helpful for blur kernel estimation. Experiments on the CelebA dataset and the GOPRO dataset show that our method performs favorably against several state-of-the-art methods.
5.Unconstrained Biometric Recognition: Summary of Recent SOCIA Lab. Research ⬇️
The development of biometric recognition solutions able to work in visual surveillance conditions, i.e., in unconstrained data acquisition conditions and under covert protocols has been motivating growing efforts from the research community. Among the various laboratories, schools and research institutes concerned about this problem, the SOCIA: Soft Computing and Image Analysis Lab., of the University of Beira Interior, Portugal, has been among the most active in pursuing disruptive solutions for obtaining such extremely ambitious kind of automata. This report summarises the research works published by elements of the SOCIA Lab. in the last decade in the scope of biometric recognition in unconstrained conditions. The idea is that it can be used as basis for someone wishing to entering in this research topic.
6.Multi-Modal Domain Adaptation for Fine-Grained Action Recognition ⬇️
Fine-grained action recognition datasets exhibit environmental bias, where multiple video sequences are captured from a limited number of environments. Training a model in one environment and deploying in another results in a drop in performance due to an unavoidable domain shift. Unsupervised Domain Adaptation (UDA) approaches have frequently utilised adversarial training between the source and target domains. However, these approaches have not explored the multi-modal nature of video within each domain. In this work we exploit the correspondence of modalities as a self-supervised alignment approach for UDA in addition to adversarial alignment.
We test our approach on three kitchens from our large-scale dataset, EPIC-Kitchens, using two modalities commonly employed for action recognition: RGB and Optical Flow. We show that multi-modal self-supervision alone improves the performance over source-only training by 2.4% on average. We then combine adversarial training with multi-modal self-supervision, showing that our approach outperforms other UDA methods by 3%.
7.A Robust Real-Time Computing-based Environment Sensing System for Intelligent Vehicle ⬇️
For intelligent vehicles, sensing the 3D environment is the first but crucial step. In this paper, we build a real-time advanced driver assistance system based on a low-power mobile platform. The system is a real-time multi-scheme integrated innovation system, which combines stereo matching algorithm with machine learning based obstacle detection approach and takes advantage of the distributed computing technology of a mobile platform with GPU and CPUs. First of all, a multi-scale fast MPV (Multi-Path-Viterbi) stereo matching algorithm is proposed, which can generate robust and accurate disparity map. Then a machine learning, which is based on fusion technology of monocular and binocular, is applied to detect the obstacles. We also advance an automatic fast calibration mechanism based on Zhang's calibration method. Finally, the distributed computing and reasonable data flow programming are applied to ensure the operational efficiency of the system. The experimental results show that the system can achieve robust and accurate real-time environment perception for intelligent vehicles, which can be directly used in the commercial real-time intelligent driving applications.
8.Explaining with Counter Visual Attributes and Examples ⬇️
In this paper, we aim to explain the decisions of neural networks by utilizing multimodal information. That is counter-intuitive attributes and counter visual examples which appear when perturbed samples are introduced. Different from previous work on interpreting decisions using saliency maps, text, or visual patches we propose to use attributes and counter-attributes, and examples and counter-examples as part of the visual explanations. When humans explain visual decisions they tend to do so by providing attributes and examples. Hence, inspired by the way of human explanations in this paper we provide attribute-based and example-based explanations. Moreover, humans also tend to explain their visual decisions by adding counter-attributes and counter-examples to explain what is not seen. We introduce directed perturbations in the examples to observe which attribute values change when classifying the examples into the counter classes. This delivers intuitive counter-attributes and counter-examples. Our experiments with both coarse and fine-grained datasets show that attributes provide discriminating and human-understandable intuitive and counter-intuitive explanations.
9.The Whole Is Greater Than the Sum of Its Nonrigid Parts ⬇️
According to Aristotle, a philosopher in Ancient Greece, "the whole is greater than the sum of its parts". This observation was adopted to explain human perception by the Gestalt psychology school of thought in the twentieth century. Here, we claim that observing part of an object which was previously acquired as a whole, one could deal with both partial matching and shape completion in a holistic manner. More specifically, given the geometry of a full, articulated object in a given pose, as well as a partial scan of the same object in a different pose, we address the problem of matching the part to the whole while simultaneously reconstructing the new pose from its partial observation. Our approach is data-driven, and takes the form of a Siamese autoencoder without the requirement of a consistent vertex labeling at inference time; as such, it can be used on unorganized point clouds as well as on triangle meshes. We demonstrate the practical effectiveness of our model in the applications of single-view deformable shape completion and dense shape correspondence, both on synthetic and real-world geometric data, where we outperform prior work on these tasks by a large margin.
10.Convolution Neural Network Architecture Learning for Remote Sensing Scene Classification ⬇️
Remote sensing image scene classification is a fundamental but challenging task in understanding remote sensing images. Recently, deep learning-based methods, especially convolutional neural network-based (CNN-based) methods have shown enormous potential to understand remote sensing images. CNN-based methods meet with success by utilizing features learned from data rather than features designed manually. The feature-learning procedure of CNN largely depends on the architecture of CNN. However, most of the architectures of CNN used for remote sensing scene classification are still designed by hand which demands a considerable amount of architecture engineering skills and domain knowledge, and it may not play CNN's maximum potential on a special dataset. In this paper, we proposed an automatically architecture learning procedure for remote sensing scene classification. We designed a parameters space in which every set of parameters represents a certain architecture of CNN (i.e., some parameters represent the type of operators used in the architecture such as convolution, pooling, no connection or identity, and the others represent the way how these operators connect). To discover the optimal set of parameters for a given dataset, we introduced a learning strategy which can allow efficient search in the architecture space by means of gradient descent. An architecture generator finally maps the set of parameters into the CNN used in our experiments.
11.Practical Fast Gradient Sign Attack against Mammographic Image Classifier ⬇️
Artificial intelligence (AI) has been a topic of major research for many years. Especially, with the emergence of deep neural network (DNN), these studies have been tremendously successful. Today machines are capable of making faster, more accurate decision than human. Thanks to the great development of machine learning (ML) techniques, ML have been used many different fields such as education, medicine, malware detection, autonomous car etc. In spite of having this degree of interest and much successful research, ML models are still vulnerable to adversarial attacks. Attackers can manipulate clean data in order to fool the ML classifiers to achieve their desire target. For instance; a benign sample can be modified as a malicious sample or a malicious one can be altered as benign while this modification can not be recognized by human observer. This can lead to many financial losses, or serious injuries, even deaths. The motivation behind this paper is that we emphasize this issue and want to raise awareness. Therefore, the security gap of mammographic image classifier against adversarial attack is demonstrated. We use mamographic images to train our model then evaluate our model performance in terms of accuracy. Later on, we poison original dataset and generate adversarial samples that missclassified by the model. We then using structural similarity index (SSIM) analyze similarity between clean images and adversarial images. Finally, we show how successful we are to misuse by using different poisoning factors.
12.FakeLocator: Robust Localization of GAN-Based Face Manipulations via Semantic Segmentation Networks with Bells and Whistles ⬇️
Nowadays, full face synthesis and partial face manipulation by virtue of the generative adversarial networks (GANs) have raised wide public concern. In the digital media forensics area, detecting and ultimately locating the image forgery have become imperative. Although many methods focus on fake detection, only a few put emphasis on the localization of the fake regions. Through analyzing the imperfection in the upsampling procedures of the GAN-based methods and recasting the fake localization problem as a modified semantic segmentation one, our proposed FakeLocator can obtain high localization accuracy, at full resolution, on manipulated facial images. To the best of our knowledge, this is the very first attempt to solve the GAN-based fake localization problem with a semantic segmentation map. As an improvement, the real-numbered segmentation map proposed by us preserves more information of fake regions. For this new type segmentation map, we also find suitable loss functions for it. Experimental results on the CelebA and FFHQ databases with seven different SOTA GAN-based face generation methods show the effectiveness of our method. Compared with the baseline, our method performs several times better on various metrics. Moreover, the proposed method is robust against various real-world facial image degradations such as JPEG compression, low-resolution, noise, and blur.
13.Crowd Scene Analysis by Output Encoding ⬇️
Crowd scene analysis receives growing attention due to its wide applications. Grasping the accurate crowd location (rather than merely crowd count) is important for spatially identifying high-risk regions in congested scenes. In this paper, we propose a Compressed Sensing based Output Encoding (CSOE) scheme, which casts detecting pixel coordinates of small objects into a task of signal regression in encoding signal space. CSOE helps to boost localization performance in circumstances where targets are highly crowded without huge scale variation. In addition, proper receptive field sizes are crucial for crowd analysis due to human size variations. We create Multiple Dilated Convolution Branches (MDCB) that offers a set of different receptive field sizes, to improve localization accuracy when objects sizes change drastically in an image. Also, we develop an Adaptive Receptive Field Weighting (ARFW) module, which further deals with scale variation issue by adaptively emphasizing informative channels that have proper receptive field size. Experiments demonstrate the effectiveness of the proposed method, which achieves state-of-the-art performance across four mainstream datasets, especially achieves excellent results in highly crowded scenes. More importantly, experiments support our insights that it is crucial to tackle target size variation issue in crowd analysis task, and casting crowd localization as regression in encoding signal space is quite effective for crowd analysis.
14.Weakly Supervised Few-shot Object Segmentation using Co-Attention with Visual and Semantic Inputs ⬇️
Significant progress has been made recently in developing few-shot object segmentation methods. Learning is shown to be successful in a few segmentation settings, including pixel-level, scribbles and bounding boxes. These methods can be classified as "strongly labelled" support images because significant image editing efforts are required to provide the labeling. This paper takes another approach, i.e., only requiring image-level classification data for few-shot object segmentation. The large amount of image-level labelled data signifies this approach, if successful. The problem is challenging because there is no obvious features that can be used for segmentation in the image-level data. We propose a novel multi-modal interaction module for few-shot object segmentation that utilizes a co-attention mechanism using both visual and word embedding. Our model using image-level labels achieves 4.8% improvement over previously proposed image-level few-shot object segmentation. It also outperforms state-of-the-art methods that use weak bounding box supervision on PASCAL-5i. Our results show that few-shot segmentation benefits from utilizing word embeddings, and that we are able to perform few-shot segmentation using stacked joint visual semantic processing with weak image-level labels. We further propose a novel setup, Temporal Object Segmentation for Few-shot Learning (TOSFL) for videos. TOSFL requires only image-level labels for the first frame in order to segment objects in the following frames. TOSFL provides a novel benchmark for video segmentation, which can be used on a variety of public video data such as Youtube-VOS, as demonstrated in our experiment.
15.Visualisation of Medical Image Fusion and Translation for Accurate Diagnosis of High Grade Gliomas ⬇️
The medical image fusion combines two or more modalities into a single view while medical image translation synthesizes new images and assists in data augmentation. Together, these methods help in faster diagnosis of high grade malignant gliomas. However, they might be untrustworthy due to which neurosurgeons demand a robust visualisation tool to verify the reliability of the fusion and translation results before they make pre-operative surgical decisions. In this paper, we propose a novel approach to compute a confidence heat map between the source-target image pair by estimating the information transfer from the source to the target image using the joint probability distribution of the two images. We evaluate several fusion and translation methods using our visualisation procedure and showcase its robustness in enabling neurosurgeons to make finer clinical decisions.
16.Using Simulated Data to Generate Images of Climate Change ⬇️
Generative adversarial networks (GANs) used in domain adaptation tasks have the ability to generate images that are both realistic and personalized, transforming an input image while maintaining its identifiable characteristics. However, they often require a large quantity of training data to produce high-quality images in a robust way, which limits their usability in cases when access to data is limited. In our paper, we explore the potential of using images from a simulated 3D environment to improve a domain adaptation task carried out by the MUNIT architecture, aiming to use the resulting images to raise awareness of the potential future impacts of climate change.
17.Abdominal multi-organ segmentation with cascaded convolutional and adversarial deep networks ⬇️
Objective : Abdominal anatomy segmentation is crucial for numerous applications from computer-assisted diagnosis to image-guided surgery. In this context, we address fully-automated multi-organ segmentation from abdominal CT and MR images using deep learning. Methods: The proposed model extends standard conditional generative adversarial networks. Additionally to the discriminator which enforces the model to create realistic organ delineations, it embeds cascaded partially pre-trained convolutional encoder-decoders as generator. Encoder fine-tuning from a large amount of non-medical images alleviates data scarcity limitations. The network is trained end-to-end to benefit from simultaneous multi-level segmentation refinements using auto-context. Results : Employed for healthy liver, kidneys and spleen segmentation, our pipeline provides promising results by outperforming state-of-the-art encoder-decoder schemes. Followed for the Combined Healthy Abdominal Organ Segmentation (CHAOS) challenge organized in conjunction with the IEEE International Symposium on Biomedical Imaging 2019, it gave us the first rank for three competition categories: liver CT, liver MR and multi-organ MR segmentation. Conclusion : Combining cascaded convolutional and adversarial networks strengthens the ability of deep learning pipelines to automatically delineate multiple abdominal organs, with good generalization capability. Significance : The comprehensive evaluation provided suggests that better guidance could be achieved to help clinicians in abdominal image interpretation and clinical decision making.
18.Unsupervised Disentanglement of Pose, Appearance and Background from Images and Videos ⬇️
Unsupervised landmark learning is the task of learning semantic keypoint-like representations without the use of expensive input keypoint-level annotations. A popular approach is to factorize an image into a pose and appearance data stream, then to reconstruct the image from the factorized components. The pose representation should capture a set of consistent and tightly localized landmarks in order to facilitate reconstruction of the input image. Ultimately, we wish for our learned landmarks to focus on the foreground object of interest. However, the reconstruction task of the entire image forces the model to allocate landmarks to model the background. This work explores the effects of factorizing the reconstruction task into separate foreground and background reconstructions, conditioning only the foreground reconstruction on the unsupervised landmarks. Our experiments demonstrate that the proposed factorization results in landmarks that are focused on the foreground object of interest. Furthermore, the rendered background quality is also improved, as the background rendering pipeline no longer requires the ill-suited landmarks to model its pose and appearance. We demonstrate this improvement in the context of the video-prediction task.
19.Brain Metastasis Segmentation Network Trained with Robustness to Annotations with Multiple False Negatives ⬇️
Deep learning has proven to be an essential tool for medical image analysis. However, the need for accurately labeled input data, often requiring time- and labor-intensive annotation by experts, is a major limitation to the use of deep learning. One solution to this challenge is to allow for use of coarse or noisy labels, which could permit more efficient and scalable labeling of images. In this work, we develop a lopsided loss function based on entropy regularization that assumes the existence of a nontrivial false negative rate in the target annotations. Starting with a carefully annotated brain metastasis lesion dataset, we simulate data with false negatives by (1) randomly censoring the annotated lesions and (2) systematically censoring the smallest lesions. The latter better models true physician error because smaller lesions are harder to notice than the larger ones. Even with a simulated false negative rate as high as 50%, applying our loss function to randomly censored data preserves maximum sensitivity at 97% of the baseline with uncensored training data, compared to just 10% for a standard loss function. For the size-based censorship, performance is restored from 17% with the current standard to 88% with our lopsided bootstrap loss. Our work will enable more efficient scaling of the image labeling process, in parallel with other approaches on creating more efficient user interfaces and tools for annotation.
20.SDOD:Real-time Segmenting and Detecting 3D Objects by Depth ⬇️
Most existing instance segmentation methods only focus on 2D objects and are not suitable for 3D scenes such as autonomous driving. In this paper, we propose a model that splits instance segmentation and object detection into two parallel branches. We discretize the objects depth into "depth categories" (background set to 0, objects set to [1, K]), then the instance segmentation task has been transformed into a pixel-level classification task. Mask branch predicts pixel-level "depth categories", 3D branch predicts instance-level "depth categories", we produce instance mask by assigning pixels which have same "depth categories" to each instance. In addition, in order to solve the problem of imbalanced between mask labels and 3D labels in the KITTI dataset (200 for mask, 7481 for 3D), we use unreal mask generated by other instance segmentation method to train mask branch. Despite the use of unreal mask labels, experiments result on KITTI dataset still achieves state-of-the-art performance in vehicle instance segmentation.
21.EEG fingerprinting: subject specific signature based on the aperiodic component of power spectrum ⬇️
During the last few years, there has been growing interest in the effects induced by individual variability on activation patterns and brain connectivity. The practical implications of individual variability is of basic relevance for both group level and subject level studies. The Electroencephalogram (EEG), still represents one of the most used recording techniques to investigate a wide range of brain related features. In this work, we aim to estimate the effect of individual variability on a set of very simple and easily interpretable features extracted from the EEG power spectra. In particular, in an identification scenario, we investigated how the aperiodic (1/f background) component of the EEG power spectra can accurately identify subjects from a large EEG dataset. The results of this study show that the aperiodic component of the EEG signal is characterized by strong subject-specific properties, that this feature is consistent across different experimental conditions (eyes-open and eyes-closed) and outperforms the canonically-defined frequency bands. These findings suggest that the simple features (slope and offset) extracted from the aperiodic component of the EEG signal are sensitive to individual traits and may help to characterize and make inferences at single subject level.
22.Curriculum Audiovisual Learning ⬇️
Associating sound and its producer in complex audiovisual scene is a challenging task, especially when we are lack of annotated training data. In this paper, we present a flexible audiovisual model that introduces a soft-clustering module as the audio and visual content detector, and regards the pervasive property of audiovisual concurrency as the latent supervision for inferring the correlation among detected contents. To ease the difficulty of audiovisual learning, we propose a novel curriculum learning strategy that trains the model from simple to complex scene. We show that such ordered learning procedure rewards the model the merits of easy training and fast convergence. Meanwhile, our audiovisual model can also provide effective unimodal representation and cross-modal alignment performance. We further deploy the well-trained model into practical audiovisual sound localization and separation task. We show that our localization model significantly outperforms existing methods, based on which we show comparable performance in sound separation without referring external visual supervision. Our video demo can be found at this https URL.
23.Scene Text Recognition With Finer Grid Rectification ⬇️
Scene Text Recognition is a challenging problem because of irregular styles and various distortions. This paper proposed an end-to-end trainable model consists of a finer rectification module and a bidirectional attentional recognition network(Firbarn). The rectification module adopts finer grid to rectify the distorted input image and the bidirectional decoder contains only one decoding layer instead of two separated one. Firbarn can be trained in a weak supervised way, only requiring the scene text images and the corresponding word labels. With the flexible rectification and the novel bidirectional decoder, the results of extensive evaluation on the standard benchmarks show Firbarn outperforms previous works, especially on irregular datasets.
24.Learning Preference-Based Similarities from Face Images using Siamese Multi-Task CNNs ⬇️
Online dating has become a common occurrence over the last few decades. A key challenge for online dating platforms is to determine suitable matches for their users. A lot of dating services rely on self-reported user traits and preferences for matching. At the same time, some services largely rely on user images and thus initial visual preference. Especially for the latter approach, previous research has attempted to capture users' visual preferences for automatic match recommendation. These approaches are mostly based on the assumption that physical attraction is the key factor for relationship formation and personal preferences, interests, and attitude are largely neglected. Deep learning approaches have shown that a variety of properties can be predicted from human faces to some degree, including age, health and even personality traits. Therefore, we investigate the feasibility of bridging image-based matching and matching with personal interests, preferences, and attitude. We approach the problem in a supervised manner by predicting similarity scores between two users based on images of their faces only. The ground-truth for the similarity matching scores is determined by a test that aims to capture users' preferences, interests, and attitude that are relevant for forming romantic relationships. The images are processed by a Siamese Multi-Task deep learning architecture. We find a statistically significant correlation between predicted and target similarity scores. Thus, our results indicate that learning similarities in terms of interests, preferences, and attitude from face images appears to be feasible to some degree.
25.Learning Canonical Shape Space for Category-Level 6D Object Pose and Size Estimation ⬇️
We present a novel approach to category-level 6D object pose and size estimation. To tackle intra-class shape variation, we learn canonical shape space (CASS), a unified representation for a large variety of instances of a certain object category. In particular, CASS is modeled as the latent space of a deep generative model of canonical 3D shapes with normalized pose and size. We train a variational auto-encoder (VAE) for generating 3D point clouds in the canonical space from an RGBD image. The VAE is trained in a cross-category fashion, exploiting the publicly available large 3D shape repositories. Since the 3D point cloud is generated in normalized pose and size, the encoder of the VAE learns view-factorized RGBD embedding. It maps an RGBD image in arbitrary view into a pose-independent 3D shape representation. Object pose and size are then estimated via contrasting it with a pose-dependent feature of the input RGBD extracted with a separate deep neural networks. We integrate the learning of CASS and pose and size estimation into an end-to-end trainable network, achieving the state-of-the-art pose and size accuracy.
26.e-UDA: Efficient Unsupervised Domain Adaptation for Cross-Site Medical Image Segmentation ⬇️
Domain adaptation in healthcare data is a potentially critical component in making computer-aided diagnostic systems applicable cross multiple sites and imaging scanners. In this paper, we propose an efficient unsupervised domain adaptation framework for robust image segmentation cross multiple similar domains. We enforce our algorithm to not only adapt to the new domains via an adversarial optimization, rejecting unlikely segmentation patterns, but also to maintain its performance on the source training data, by incorporating both semantic and boundary information into the data distributions. Further, as we do not have labels for the transfer domain, we propose a new quality score for the adaptation process, and strategies to retrain the diagnostic algorithm in a stable fashion. Using multi-centric data from a public benchmark for brain lesion segmentation, we demonstrate that recalibrating on just few unlabeled image sets from the target domain improves segmentation accuracies drastically, with performances almost similar to those from algorithms trained on fresh and fully annotated data from the test domain.
27.Look Closer to Ground Better: Weakly-Supervised Temporal Grounding of Sentence in Video ⬇️
In this paper, we study the problem of weakly-supervised temporal grounding of sentence in video. Specifically, given an untrimmed video and a query sentence, our goal is to localize a temporal segment in the video that semantically corresponds to the query sentence, with no reliance on any temporal annotation during training. We propose a two-stage model to tackle this problem in a coarse-to-fine manner. In the coarse stage, we first generate a set of fixed-length temporal proposals using multi-scale sliding windows, and match their visual features against the sentence features to identify the best-matched proposal as a coarse grounding result. In the fine stage, we perform a fine-grained matching between the visual features of the frames in the best-matched proposal and the sentence features to locate the precise frame boundary of the fine grounding result. Comprehensive experiments on the ActivityNet Captions dataset and the Charades-STA dataset demonstrate that our two-stage model achieves compelling performance.
28.GEDDnet: A Network for Gaze Estimation with Dilation and Decomposition ⬇️
Appearance-based gaze estimation from RGB images provides relatively unconstrained gaze tracking from commonly available hardware. The accuracy of subject-independent models is limited partly by small intra-subject and large inter-subject variations in appearance, and partly by a latent subject-dependent bias. To improve estimation accuracy, we propose to use dilated-convolutions in a deep convolutional neural network to capture subtle changes in the eye images, and a novel gaze decomposition method that decomposes the gaze angle into the sum of a subject-independent gaze estimate from the image and a subject-dependent bias. To further reduce estimation error, we propose a calibration method that estimates the bias from a few images taken as the subject gazes at only a few or even just a single gaze target. This significantly redues calibration time and complexity. Experiments on four datasets, including a new dataset we collected containing large variations in head pose and face location, indicate that even without calibration the estimator already outperforms state-of-the-art methods by more than 6.3%. The proposed calibration method is robust to the location of calibration target and reduces estimation error significantly (up to 35.6%), achieving state-of-the-art performance with much less calibration data than required by previously proposed methods.
29.On the Role of Receptive Field in Unsupervised Sim-to-Real Image Translation ⬇️
Generative Adversarial Networks (GANs) are now widely used for photo-realistic image synthesis. In applications where a simulated image needs to be translated into a realistic image (sim-to-real), GANs trained on unpaired data from the two domains are susceptible to failure in semantic content retention as the image is translated from one domain to the other. This failure mode is more pronounced in cases where the real data lacks content diversity, resulting in a content \emph{mismatch} between the two domains - a situation often encountered in real-world deployment. In this paper, we investigate the role of the discriminator's receptive field in GANs for unsupervised image-to-image translation with mismatched data, and study its effect on semantic content retention. Experiments with the discriminator architecture of a state-of-the-art coupled Variational Auto-Encoder (VAE) - GAN model on diverse, mismatched datasets show that the discriminator receptive field is directly correlated with semantic content discrepancy of the generated image.
30.PSC-Net: Learning Part Spatial Co-occurence for Occluded Pedestrian Detection ⬇️
Detecting pedestrians, especially under heavy occlusions, is a challenging computer vision problem with numerous real-world applications. This paper introduces a novel approach, termed as PSC-Net, for occluded pedestrian detection. The proposed PSC-Net contains a dedicated module that is designed to explicitly capture both inter and intra-part co-occurrence information of different pedestrian body parts through a Graph Convolutional Network (GCN). Both inter and intra-part co-occurrence information contribute towards improving the feature representation for handling varying level of occlusions, ranging from partial to severe occlusions. Our PSC-Net exploits the topological structure of pedestrian and does not require part-based annotations or additional visible bounding-box (VBB) information to learn part spatial co-occurence. Comprehensive experiments are performed on two challenging datasets: CityPersons and Caltech datasets. The proposed PSC-Net achieives state-of-the-art detection performance on both. On the heavy occluded (HO) set of CityPerosns test set, our PSC-Net obtains an absolute gain of 3.6% in terms of log-average miss rate over the state-of-the-art with same backbone, input scale and without using additional VBB supervision. Further, PSC-Net improves the state-of-the-art from 37.9 to 34.9 in terms of log-average miss rate on Caltech (\textbf{HO}) test set.
31.Temporal Pulses Driven Spiking Neural Network for Fast Object Recognition in Autonomous Driving ⬇️
Accurate real-time object recognition from sensory data has long been a crucial and challenging task for autonomous driving. Even though deep neural networks (DNNs) have been successfully applied in this area, most existing methods still heavily rely on the pre-processing of the pulse signals derived from LiDAR sensors, and therefore introduce additional computational overhead and considerable latency. In this paper, we propose an approach to address the object recognition problem directly with raw temporal pulses utilizing the spiking neural network (SNN). Being evaluated on various datasets (including Sim LiDAR, KITTI and DVS-barrel) derived from LiDAR and dynamic vision sensor (DVS), our proposed method has shown comparable performance as the state-of-the-art methods, while achieving remarkable time efficiency. It highlights the SNN's great potentials in autonomous driving and related applications. To the best of our knowledge, this is the first attempt to use SNN to directly perform object recognition on raw temporal pulses.
32.Modular network for high accuracy object detection ⬇️
We present a novel modular object detection convolutional neural network that significantly improves the accuracy of computer vision object detection. The network consists of two stages in a hierarchical structure. The first stage is a network that detects general classes. The second stage consists of separate networks to refine the classification and localization of each of the general classes objects. Compared to a state of the art object detection networks the classification error in the modular network is improved by approximately 3-5 times, from 12 percent to 2.5-4.5 percent. The modular network achieved a very high score in object detection of 0.94 mAP. The network is easy to implement, it can be a platform to improve the accuracy of widespread state of the art object detection networks and other kinds of deep learning networks.
33.VerSe: A Vertebrae Labelling and Segmentation Benchmark ⬇️
In this paper we report the challenge set-up and results of the Large Scale Vertebrae Segmentation Challenge (VerSe) organized in conjunction with the MICCAI 2019. The challenge consisted of two tasks, vertebrae labelling and vertebrae segmentation. For this a total of 160 multidetector CT scan cohort closely resembling clinical setting was prepared and was annotated at a voxel-level by a human-machine hybrid algorithm. In this paper we also present the annotation protocol and the algorithm that aided the medical experts in the annotation process. Eleven fully automated algorithms were benchmarked on this data with the best performing algorithm achieving a vertebrae identification rate of 95% and a Dice coefficient of 90%. VerSe'19 is an open-call challenge at its image data along with the annotations and evaluation tools will continue to be publicly accessible through its online portal.
34.Learning a distance function with a Siamese network to localize anomalies in videos ⬇️
This work introduces a new approach to localize anomalies in surveillance video. The main novelty is the idea of using a Siamese convolutional neural network (CNN) to learn a distance function between a pair of video patches (spatio-temporal regions of video). The learned distance function, which is not specific to the target video, is used to measure the distance between each video patch in the testing video and the video patches found in normal training video. If a testing video patch is not similar to any normal video patch then it must be anomalous. We compare our approach to previously published algorithms using 4 evaluation measures and 3 challenging target benchmark datasets. Experiments show that our approach either surpasses or performs comparably to current state-of-the-art methods.
35.End-to-End Vision-Based Adaptive Cruise Control (ACC) Using Deep Reinforcement Learning ⬇️
This paper presented a deep reinforcement learning method named Double Deep Q-networks to design an end-to-end vision-based adaptive cruise control (ACC) system. A simulation environment of a highway scene was set up in Unity, which is a game engine that provided both physical models of vehicles and feature data for training and testing. Well-designed reward functions associated with the following distance and throttle/brake force were implemented in the reinforcement learning model for both internal combustion engine (ICE) vehicles and electric vehicles (EV) to perform adaptive cruise control. The gap statistics and total energy consumption are evaluated for different vehicle types to explore the relationship between reward functions and powertrain characteristics. Compared with the traditional radar-based ACC systems or human-in-the-loop simulation, the proposed vision-based ACC system can generate either a better gap regulated trajectory or a smoother speed trajectory depending on the preset reward function. The proposed system can be well adaptive to different speed trajectories of the preceding vehicle and operated in real-time.
36.Weakly Supervised Lesion Co-segmentation on CT Scans ⬇️
Lesion segmentation in medical imaging serves as an effective tool for assessing tumor sizes and monitoring changes in growth. However, not only is manual lesion segmentation time-consuming, but it is also expensive and requires expert radiologist knowledge. Therefore many hospitals rely on a loose substitute called response evaluation criteria in solid tumors (RECIST). Although these annotations are far from precise, they are widely used throughout hospitals and are found in their picture archiving and communication systems (PACS). Therefore, these annotations have the potential to serve as a robust yet challenging means of weak supervision for training full lesion segmentation models. In this work, we propose a weakly-supervised co-segmentation model that first generates pseudo-masks from the RECIST slices and uses these as training labels for an attention-based convolutional neural network capable of segmenting common lesions from a pair of CT scans. To validate and test the model, we utilize the DeepLesion dataset, an extensive CT-scan lesion dataset that contains 32,735 PACS bookmarked images. Extensive experimental results demonstrate the efficacy of our co-segmentation approach for lesion segmentation with a mean Dice coefficient of 90.3%.
37.Rotation, Translation, and Cropping for Zero-Shot Generalization ⬇️
Deep Reinforcement Learning (DRL) has shown impressive performance on domains with visual inputs, in particular various games. However, the agent is usually trained on a fixed environment, e.g. a fixed number of levels. A growing mass of evidence suggests that these trained models fail to generalize to even slight variations of the environments they were trained on. This paper advances the hypothesis that the lack of generalization is partly due to the input representation, and explores how rotation, cropping and translation could increase generality. We show that a cropped, translated and rotated observation can get better generalization on unseen levels of a two-dimensional arcade game. The generality of the agent is evaluated on a set of human-designed levels.
38.Genetic Programming for Evolving a Front of Interpretable Models for Data Visualisation ⬇️
Data visualisation is a key tool in data mining for understanding big datasets. Many visualisation methods have been proposed, including the well-regarded state-of-the-art method t-Distributed Stochastic Neighbour Embedding. However, the most powerful visualisation methods have a significant limitation: the manner in which they create their visualisation from the original features of the dataset is completely opaque. Many domains require an understanding of the data in terms of the original features; there is hence a need for powerful visualisation methods which use understandable models. In this work, we propose a genetic programming approach named GPtSNE for evolving interpretable mappings from a dataset to highquality visualisations. A multi-objective approach is designed that produces a variety of visualisations in a single run which give different trade-offs between visual quality and model complexity. Testing against baseline methods on a variety of datasets shows the clear potential of GP-tSNE to allow deeper insight into data than that provided by existing visualisation methods. We further highlight the benefits of a multi-objective approach through an in-depth analysis of a candidate front, which shows how multiple models can
39.aiTPR: Attribute Interaction-Tensor Product Representation for Image Caption ⬇️
Region visual features enhance the generative capability of the machines based on features, however they lack proper interaction attentional perceptions and thus ends up with biased or uncorrelated sentences or pieces of misinformation. In this work, we propose Attribute Interaction-Tensor Product Representation (aiTPR) which is a convenient way of gathering more information through orthogonal combination and learning the interactions as physical entities (tensors) and improving the captions. Compared to previous works, where features are added up to undefined feature spaces, TPR helps in maintaining sanity in combinations and orthogonality helps in defining familiar spaces. We have introduced a new concept layer that defines the objects and also their interactions that can play a crucial role in determination of different descriptions. The interaction portions have contributed heavily for better caption quality and has out-performed different previous works on this domain and MSCOCO dataset. We introduced, for the first time, the notion of combining regional image features and abstracted interaction likelihood embedding for image captioning.
40.Imperfect ImaGANation: Implications of GANs Exacerbating Biases on Facial Data Augmentation and Snapchat Selfie Lenses ⬇️
Recently, the use of synthetic data generated by GANs has become a popular method to do data augmentation for many applications. While practitioners celebrate this as an economical way to obtain synthetic data for training data-hungry machine learning models, it is not clear that they recognize the perils of such an augmentation technique when applied to an already-biased dataset. Although one expects GANs to replicate the distribution of the original data, in real-world settings with limited data and finite network capacity, GANs suffer from mode collapse. Especially when this data is coming from online social media platforms or the web which are never balanced. In this paper, we show that in settings where data exhibits bias along some axes (eg. gender, race), failure modes of Generative Adversarial Networks (GANs) exacerbate the biases in the generated data. More often than not, this bias is unavoidable; we empirically demonstrate that given input of a dataset of headshots of engineering faculty collected from 47 online university directory webpages in the United States is biased toward white males, a state-of-the-art (unconditional variant of) GAN "imagines" faces of synthetic engineering professors that have masculine facial features and white skin color (inferred using human studies and a state-of-the-art gender recognition system). We also conduct a preliminary case study to highlight how Snapchat's explosively popular "female" filter (widely accepted to use a conditional variant of GAN), ends up consistently lightening the skin tones in women of color when trying to make face images appear more feminine. Our study is meant to serve as a cautionary tale for the lay practitioners who may unknowingly increase the bias in their training data by using GAN-based augmentation techniques with web data and to showcase the dangers of using biased datasets for facial applications.
41.Markov-Chain Monte Carlo Approximation of the Ideal Observer using Generative Adversarial Networks ⬇️
The Ideal Observer (IO) performance has been advocated when optimizing medical imaging systems for signal detection tasks. However, analytical computation of the IO test statistic is generally intractable. To approximate the IO test statistic, sampling-based methods that employ Markov-Chain Monte Carlo (MCMC) techniques have been developed. However, current applications of MCMC techniques have been limited to several object models such as a lumpy object model and a binary texture model, and it remains unclear how MCMC methods can be implemented with other more sophisticated object models. Deep learning methods that employ generative adversarial networks (GANs) hold great promise to learn stochastic object models (SOMs) from image data. In this study, we described a method to approximate the IO by applying MCMC techniques to SOMs learned by use of GANs. The proposed method can be employed with arbitrary object models that can be learned by use of GANs, thereby the domain of applicability of MCMC techniques for approximating the IO performance is extended. In this study, both signal-known-exactly (SKE) and signal-known-statistically (SKS) binary signal detection tasks are considered. The IO performance computed by the proposed method is compared to that computed by the conventional MCMC method. The advantages of the proposed method are discussed.
42.Explainable Artificial Intelligence and Machine Learning: A reality rooted perspective ⬇️
We are used to the availability of big data generated in nearly all fields of science as a consequence of technological progress. However, the analysis of such data possess vast challenges. One of these relates to the explainability of artificial intelligence (AI) or machine learning methods. Currently, many of such methods are non-transparent with respect to their working mechanism and for this reason are called black box models, most notably deep learning methods. However, it has been realized that this constitutes severe problems for a number of fields including the health sciences and criminal justice and arguments have been brought forward in favor of an explainable AI. In this paper, we do not assume the usual perspective presenting explainable AI as it should be, but rather we provide a discussion what explainable AI can be. The difference is that we do not present wishful thinking but reality grounded properties in relation to a scientific theory beyond physics.
43.Deep Learning-based Image Compression with Trellis Coded Quantization ⬇️
Recently many works attempt to develop image compression models based on deep learning architectures, where the uniform scalar quantizer (SQ) is commonly applied to the feature maps between the encoder and decoder. In this paper, we propose to incorporate trellis coded quantizer (TCQ) into a deep learning based image compression framework. A soft-to-hard strategy is applied to allow for back propagation during training. We develop a simple image compression model that consists of three subnetworks (encoder, decoder and entropy estimation), and optimize all of the components in an end-to-end manner. We experiment on two high resolution image datasets and both show that our model can achieve superior performance at low bit rates. We also show the comparisons between TCQ and SQ based on our proposed baseline model and demonstrate the advantage of TCQ.