Morphological hierarchies form a popular framework aiming at emphasizing the multiscale structure of digital image by performing an unsupervised spatial partitioning of the data. These hierarchies have been recently extended to cope with image sequences, and different strategies have been proposed to allow their construction from spatio-temporal data. In this paper, we compare these hierarchical representation strategies for image sequences according to their structural properties. We introduce a projection method to make these representations comparable. Furthermore, we extend one of these recent strategies in order to obtain more efficient hierarchical representations for image sequences. Experiments were conducted on both synthetic and real datasets, the latter being made of satellite image time series. We show that building one hierarchy by using spatial and temporal information together is more efficient comparing to other existing strategies.

Malware classification helps to understand its purpose and is also an important part of attack detection. And it is also an important part of discovering attacks. Due to continuous innovation and development of artificial intelligence, it is a trend to combine deep learning with malware classification. In this paper, we propose an improved malware image rescaling algorithm (IMIR) based on local mean algorithm. Its main goal of IMIR is to reduce the loss of information from samples during the process of converting binary files to image files. Therefore, we construct a neural network structure based on VGG model, which is suitable for image classification. In the real world, a mass of malware family labels are inaccurate or lacking. To deal with this situation, we propose a novel method to train the deep neural network by Semi-supervised Generative Adversarial Network (SGAN), which only needs a small amount of malware that have accurate labels about families. By integrating SGAN with weak coupling, we can retain the weak links of supervised part and unsupervised part of SGAN. It improves the accuracy of malware classification by making classifiers more independent of discriminators. The results of experimental demonstrate that our model achieves exhibiting favorable performance. The recalls of each family in our data set are all higher than 93.75%.

The detection of logos in images, for instance, logos of airlines on airplane tails, is a difficult task in real-world weather conditions. Most systems used for logo detection are very good at detecting logos in clean images. However, they exhibit problems when images are degraded by effects of adverse weather conditions as they frequently occur in real-world scenarios. For investigating this problem on airline logo detection as a sub-problem of logo detection, we first present a new dataset for airline logo detection on airplane tails containing a test split with images degraded by adverse weather effects. Second, to handle the detection of airline logos effectively, a new two-stage airline logo detection system based on a state-of-the-art object proposal generation system and a specifically tailored classifier is proposed. Finally, improving the results on images degraded by adverse weather effects, we introduce a learning-free application-agnostic data augmentation strategy simulating effects like rain and fog. The results show the superior performance of our airline logo detection system compared to state-of-the-art. Furthermore, applying our data augmentation approach to a variety of systems, reduces the significant drop in performance on degraded images.

Few-shot classification aims to learn a well performance model with very limited labeled examples. There are mainly two directions for this aim, namely, meta- and metric-learning. Meta learning trains models in a particular way to fast adapt to new tasks, but it neglects variational features of images. Metric learning considers relationships among same or different classes, however on the downside, it usually fails to achieve competitive performance on unseen boundary examples. In this paper, we propose a Meta Generalized Network (MGNet) that aims to combine advantages of both meta- and metric-learning. There are two novel components in MGNet. Specifically, we first develop a meta backbone training method that both learns a flexible feature extractor and a classifier initializer efficiently, delightedly leading to fast adaption to unseen few-shot tasks without overfitting. Second, we design a trainable adaptive interval model to improve the cosine classifier, which increases the recognition accuracy of hard examples. We train the meta backbone in the training stage by all classes, and fine-tune the meta-backbone as well as train the adaptive classifier in the testing stage.

Automatic facial expression recognition (FER) has gained much attention due to its applications in human-computer interaction. Among the approaches to improve FER tasks, this paper focuses on deep architecture with the attention mechanism. We propose a novel Masking idea to boost the performance of CNN in facial expression task. It uses a segmentation network to refine feature maps, enabling the network to focus on relevant information to make correct decisions. In experiments, we combine the ubiquitous Deep Residual Network and Unet-like architecture to produce a Residual Masking Network. The proposed method holds state-of-the-art (SOTA) accuracy on the well-known FER2013 and private VEMO datasets. Our works are available on Github.

With the rise of the so-called Adversarial Attacks, there is an increased concern on model security. In this paper we present two different contributions: novel measures of robustness (based on adversarial attacks) and a novel adversarial attack. The key idea behind these metrics is to obtain a measure that could compare different architectures, with independence of how the input is preprocessed (robustness against different input sizes and value ranges). To do so, a novel adversarial attack is presented, performing a delayed elastic-net adversarial attack (constraints are only used whenever a successful adversarial attack is obtained). Experimental results show that our approach obtains state-of-the-art adversarial samples, in terms of minimal perturbation distance. Finally, a benchmark of ImageNet pretrained models is used to conduct experiments aiming to shed some light about which model should be selected whenever security is a role factor.

Batch Normalization (BN) is a well-known technique used in training deep neural networks. The main idea behind batch normalization is to normalize the features of the layers ($i.e.$, transforming them to have a mean equal to zero and a variance equal to one). Such a procedure encourages the optimization landscape of the loss function to be smoother, and improve the learning of the networks for both speed and performance. In this paper, we demonstrate that the performance of the network can be improved, if the distributions of the features of the output in the same layer are similar. As normalizing based on mean and variance does not necessarily make the features to have the same distribution, we propose a new normalization scheme: Batch Normalization with Skewness Reduction (BNSR). Comparing with other normalization approaches, BNSR transforms not just only the mean and variance, but also the skewness of the data. By tackling this property of a distribution, we are able to make the output distributions of the layers to be further similar. The nonlinearity of BNSR may further improve the expressiveness of the underlying network. Comparisons with other normalization schemes are tested on the CIFAR-100 and ImageNet datasets. Experimental results show that the proposed approach can outperform other state-of-the-arts that are not equipped with BNSR.

Dense pixel matching is required for many computer vision algorithms such as disparity, optical flow or scene flow estimation. Feature Pyramid Networks (FPN) have proven to be a suitable feature extractor for CNN-based dense matching tasks. FPN generates well localized and semantically strong features at multiple scales. However, the generic FPN is not utilizing its full potential, due to its reasonable but limited localization accuracy. Thus, we present ResFPN – a multiresolution feature pyramid network with multiple residual skip connections, where at any scale, we leverage the information from higher resolution maps for stronger and better localized features. In our ablation study we demonstrate the effectiveness of our novel architecture with clearly higher accuracy than FPN. In addition, we verify the superior accuracy of ResFPN in many different pixel matching applications on established datasets like KITTI, Sintel, and FlyingThings3D.

In Continual Learning, a Neural Network is trained on a stream of data whose distribution shifts over time. Under these assumptions, it is especially challenging to improve on classes appearing later in the stream while remaining accurate on previous ones. This is due to the infamous problem of catastrophic forgetting, which causes a quick performance degradation when the classifier focuses on learning new categories. Recent literature proposed various approaches to tackle this issue, often resorting to very sophisticated techniques. In this work, we show that naive rehearsal can be patched to achieve similar performance. We point out some shortcomings that restrain Experience Replay (ER) and propose five tricks to mitigate them. Experiments show that ER, thus enhanced, displays an accuracy gain of 51.2 and 26.9 percentage points on the CIFAR-10 and CIFAR-100 datasets respectively (memory buffer size 1000). As a result, it surpasses current state-of-the-art rehearsal-based methods.

In recent years, surveillance cameras are widely deployed in public places, and the general crime rate has been reduced significantly due to these ubiquitous devices. Usually, these cameras provide cues and evidence after crimes were conducted, while they are rarely used to prevent or stop criminal activities in time. It is both time and labor consuming to manually monitor a large amount of video data from surveillance cameras. Therefore, automatically recognizing violent behaviors from video signals becomes essential. In this paper, we summarize several existing video datasets for violence detection and propose a new video dataset with 2,000 videos all captured by surveillance cameras in real-world scenes. Also, we present a new method that utilizes both the merits of 3D-CNNs and optical flow, namely Flow Gated Network. The proposed approach obtains an accuracy of 87.25% on the test set of our proposed RWF-2000 database. The proposed database and source codes of this paper are currently open to access.

Detecting and segmenting individual cells from microscopy images is critical to various life science applications. Traditional cell segmentation tools are often ill-suited for applications in brightfield microscopy due to poor contrast and intensity heterogeneity, and only a small subset are applicable to segment cells in a cluster. In this regard, we introduce a novel supervised technique for cell segmentation in a multi-task learning paradigm. A combination of a multi-task loss, based on the region and cell boundary detection, is employed for an improved prediction efficiency of the network. The learning problem is posed in a novel min-max framework which enables adaptive estimation of the hyper-parameters in an automatic fashion. The region and cell boundary predictions are combined via morphological operations and active contour model to segment individual cells. The proposed methodology is particularly suited to segment touching cells from brightfield microscopy images without manual interventions. Quantitatively, we observe an overall Dice score of 0.93 on the validation set, which is an improvement of over 15.9% on a recent unsupervised method, and outperforms the popular supervised U-net algorithm by at least 5.8% on average.

In this paper, we propose a novel framework for class agnostic co-segmentation of multiple images using comparatively smaller datasets. We have developed a novel encoder-decoder network termed as DVICE (Directed Variational Inference Cross Encoder), which learns a continuous embedding space to ensure better similarity learning. We employ a combination of the proposed variational encoder-decoder and a novel few-shot learning approach to tackle the small sample size problem in co-segmentation. Furthermore, the proposed framework does not use any semantic class labels and is entirely class agnostic. Through exhaustive experimentation using a small volume of data over multiple datasets, we have demonstrated that our approach outperforms all existing state-of-the-art techniques.

Identity-preserved face beauty transformation aims to change the beauty scale of a face image while preserving the identity of the original face. In our framework of conditional Generative Adversarial Networks (cGANs), the synthesized face produced by the generator would have the same beauty scale indicated by the input condition. Unlike the discrete class labels used in most cGANs, the condition of target beauty scale in our framework is given by a continuous real-valued beauty score in the range [1 to 5], which makes the work challenging. To tackle the problem, we have implemented a triple structure, in which the conditional discriminator is divided into a normal discriminator and a separate face beauty predictor. We have also developed another new structure called Conditioned Instance Normalization to replace the original concatenation used in cGANs, which makes the combination of the input image and condition more effective. Furthermore, Self-Consistency Loss is introduced as a new parameter to improve the stability of training and quality of the generated image. In the end, the objectives of beauty transformation and identity preservation are evaluated by the pretrained face beauty predictor and state-of-the-art face recognition network. The result is encouraging and it also shows that certain facial features could be synthesized by the generator according to the target beauty scale, while preserving the original identity.

Explainable machine learning has become increasingly prevalent, especially in healthcare where explainable models are vital for ethical and trusted automated decision making. Work on the susceptibility of deep learning models to adversarial attacks has shown the ease of designing samples to mislead a model into making incorrect predictions. In this work, we propose an explainability-based method for the accurate detection of adversarial samples on two datasets with different complexity and properties: Electronic Health Record (EHR) and chest X-ray (CXR) data. On the MIMIC-III and Henan-Renmin EHR datasets, we report a detection accuracy of 77% against the Longitudinal Adversarial Attack. On the MIMIC-CXR dataset, we achieve an accuracy of 88%; significantly improving on the state of the art of adversarial detection in both datasets by over 10% in all settings. We propose an anomaly detection based method using explainability techniques to detect adversarial samples which is able to generalise to different attack methods without a need for retraining.

Representing the brain activities by networks is very crucial to understand the human brain and design brain-computer interfaces. This study proposes a novel method to estimate static and dynamic brain networks using Kulback- Leibler divergence. The suggested brain networks are based on the probability distributions of voxel intensity values measured by functional Magnetic Resonance Images (fMRI) recorded while the subjects perform a predefined cognitive task, namely complex problem-solving. We investigate the validity of the estimated brain networks by modeling and analyzing the different phases of the complex problem-solving processes of the human brain, namely planning and execution phases. The suggested computational network model is tested by Support Vector Machines. We observe that the network models can successfully discriminate the planning and execution phases of the complex problem-solving process with more than 90% accuracy, when the suggested dynamic networks, extracted from the fMRI data, are classified by Support Vector Machines.

Moreover, the abstraction of audio features makes it impossible to fully use the inherent relationship among audio features. This paper proposes a model that combines a convolutional neural network(CNN) and a capsule neural network (CapsNet), named as CapCNN. The advantage of CapCNN lies in that it provides a solution to solve time sensitivity and focus on the overall characteristics. In this study, it is found that CapCNN can well handle the speech emotion recognition task. Compared with other state-of-art methods, our algorithm shows high performances on the CASIA and EMODB datasets. The detailed analysis confirms that our method provides balanced results on the various classes.

As a typical cross-modal problem, visual question answering (VQA) has received increasing attention from the communities of computer vision and natural language processing. Reading and reasoning about texts and visual contents in the images is a burgeoning and important research topic in VQA, especially for the visually impaired assistance applications. Given an image, it aims to predict an answer to a provided natural language question closely related to its textual contents. In this paper, we propose a novel end-to-end textual content based VQA model, which grounds question answering both on the visual and textual information. After encoding the image, question and recognized text words, it uses multi-modal factorized high-order modules and the attention mechanism to fuse question-image and question-text features respectively. The complex correlations among different features can be captured efficiently. To ensure the model's extendibility, it embeds candidate answers and recognized texts in a semantic embedding space and adopts semantic embedding based classifier to perform answer prediction. Extensive experiments on the newly proposed benchmark TextVQA demonstrate that the proposed model can achieve promising results.

We present a multimodal deep learning framework that can generate summarization text supporting the main idea of an information graphic for presentation to a person who is blind or visually impaired. The framework utilizes the visual, textual, positional, and size characteristics extracted from the image to create the summary. Different and complimentary neural architectures are optimized for each task using crowdsourced training data. From our quantitative experiments and results, we explain the reasoning behind our framework and show the effectiveness of our models. Our qualitative results showcase text generated from our framework and show that Mechanical Turk participants favor them to other automatic and human generated summarizations. We describe the design and of of an experiment to evaluate the utility of our system for people who have visual impairments in the context of understanding Twitter Tweets containing line graphs.

Learning-based methods have enabled the recovery of a video sequence from a single motion-blurred image or a single coded exposure image. Recovering video from a single motion-blurred image is a very ill-posed problem and the recovered video usually has many artifacts. In addition to this, the direction of motion is lost and it results in motion ambiguity. However, it has the advantage of fully preserving the information in the static parts of the scene. The traditional coded exposure framework is better-posed but it only samples a fraction of the space-time volume, which is at best $50\%$ of the space-time volume. Here, we propose to use the complementary information present in the fully-exposed (blurred) image along with the coded exposure image to recover a high fidelity video without any motion ambiguity. Our framework consists of a shared encoder followed by an attention module to selectively combine the spatial information from the fully-exposed image with the temporal information from the coded image, which is then super-resolved to recover a non-ambiguous high-quality video. The input to our algorithm is a fully-exposed and coded image pair. Such an acquisition system already exists in the form of a Coded-two-bucket (C2B) camera. We demonstrate that our proposed deep learning approach using blurred-coded image pair produces much better results than those from just a blurred image or just a coded image.

We design a video analytics system to measure gait over time and detect trend and outliers in the data. The purpose is for health monitoring, the thesis being that trend especially can lead to early detection of declining health and be used to prevent accidents such as falls in the elderly. We use the OpenPose deep learning tool for recognizing the back and neck angle features of walking people, and measure speed as well. Trend and outlier statistics are calculated upon time series of these features. A challenge in this work is lack of testing data of decaying gait. We first designed experiments to measure consistency of the system on a healthy population, then analytically altered this real data to simulate gait decay. Results on about 4000 gait samples of 50 people over 3 months showed good separation of healthy gait subjects from those with trend or outliers, and furthermore the trend measurement was able to detect subtle decay in gait not easily discerned by the human eye.

Activity recognition in children and infants is important in applications such as safety monitoring, behavior assessment, and child-robot interaction, among others. However, it differs from activity recognition in adults not only because body poses and proportions are different, but also because of the way in which actions are performed. This paper addresses the problem of infant action classification (up to 2 years old) in challenging conditions. The actions are performed in a pediatric rehabilitation environment in which not only infants but also robots and adults are present, with the infant being one of the smallest actors in the scene. We propose a multiview action classification system based on Faster R-CNN and LSTM networks, which fuses information from different views by using learnable fusion coefficients derived from detection confidence scores. The proposed system is view-independent, learns features that are close to view-invariant, and can handle new or missing views at test time. Our approach outperforms the state-of-the-art baseline model for this dataset by 11.4% in terms of average classification accuracy in four classes (crawl, sit, stand and walk). Moreover, experiments in a extended dataset from 6 subjects (8 to 24 months old) show that the proposed fusion strategy outperforms the best post-processing fusion strategy by 2.5% and 6.8% average classification accuracy in Leave One Super-session Out and Leave One Subject Out cross-validation, respectively.