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.

Human-object interaction recognition aims for identifying the relationship between a human subject and an object. Researchers incorporate global scene context into the early layers of deep Convolutional Neural Networks as a solution. They report a significant increase in the performance since generally interactions are correlated with the scene (i.e. riding bicycle on the city street). However, this approach leads to the following problems. It increases the network size in the early layers, therefore not efficient. It leads to noisy filter responses when the scene is irrelevant, therefore not accurate. It only leverages scene context whereas human-object interactions offer a multitude of contexts, therefore incomplete. To circumvent these issues, in this work, we propose Self-Selective Context (SSC). SSC operates on the joint appearance of human-objects and context to bring the most discriminative context(s) into play for recognition. We devise novel contextual features that model the locality of human-object interactions and show that SSC can seamlessly integrate with the State-of-the-art interaction recognition models. Our experiments show that SSC leads to an important increase in interaction recognition performance, while using much fewer parameters.

Objects and activities (temporal objects) are composed of a recursive hierarchy of perceptual wholes and parts, whose properties, such as shape, reflectance, and color, constitute a hierarchy of intrinsic causal factors of object appearance.However, object appearance is the compositional consequence of both an object’s intrinsic and extrinsic causal factors, where the extrinsic causal factors are related to illumination, and imaging conditions.We propose a unified tensor model of wholes and parts that statistically models the mechanism of data formation. We derive a compositional hierarchical block tensor factorization that computes a disentangled representation of the causal factors of object image formation by optimizing simultaneously across wholes and parts. Given computational efficiency considerations,we introduce an incremental bottom-up computational alternative that employs the lower level part representations to represent the higher level of abstractions, the parent wholes. This incremental computational approach may also be employed to update the causal model representation when data that becomes available incrementally. The resulting object representation is an interpretable combinatorial choice of intrinsic causal factor representations related to an object’s recursive hierarchy of wholes and parts that renders object recognition robust to occlusion and reduces training data requirements.

Typically, real-world requirements to deploy face recognition models in unconstrained surveillance scenarios demand to identify low-resolution faces with extremely low computational cost. In the last years, several methods based on complex deep learning models have been proposed with promising recognition results but at a high computational cost. Inspired by the compactness and computation efficiency of lightweight deep face networks and their high accuracy on general face recognition tasks, in this work we propose to benchmark two recently introduced lightweight face models on low-resolution surveillance imagery to enable efficient system deployment. In this way, we conduct a comprehensive evaluation on the two typical settings: LR-to-HR and LR-to-LR matching. In addition, we investigate the effect of using trained models with down-sampled synthetic data from high-resolution images, as well as the combination of different models, for face recognition on real low-resolution images. Experimental results show that the used lightweight face models achieve state-of-the-art results on low-resolution benchmarks with low memory footprint and computational complexity. Moreover, we observed that combining models trained with different degradations improves the recognition accuracy on low-resolution surveillance imagery, which is feasible due to their low computational cost.

Localizing page elements/objects such as tables, figures, equations, etc. is the primary step in extracting information from document images. We propose a novel end-to-end trainable deep network, (CDeC-Net) for detecting tables present in the documents. The proposed network consists of a multistage extension of Mask R-CNN with a dual backbone having deformable convolution for detecting tables varying in scale with high detection accuracy at higher IoU threshold. We empirically evaluate CDeC-Net on all the publicly available benchmark datasets— ICDAR-2013, ICDAR-2017, ICDAR-2019, UNLV, Marmot, PubLayNet, TableBank, and IIIT-AR-13K —with extensive experiments. Our solution has three important properties:(i) a single trained model CDeC-Net‡ performs well across all the popular benchmark datasets; (ii) we report excellent performances across multiple, including higher, thresholds of IoU; (iii) by following the same protocol of the recent papers for each of the benchmarks, we consistently demonstrate the superior quantitative performance. Our code and models will be publicly released for enabling reproducibility of the results.

For image ranking tasks with naturally continuous output, such as age and scenicness estimation, it is common to discretize the label range and apply methods from (ordered) classification analysis. In this paper, we propose a data-driven approach for simultaneous representation learning and label discretization. Compared to arbitrarily selecting thresholds, we seek to learn thresholds and image representations by minimizing a novel loss function in an end-to-end model. We demonstrate our combined approach on a variety of image ranking tasks and demonstrate that it outperforms task-specific methods. Additionally, our learned partitioning scheme can be transferred to improve methods that rely on discretization.

Creating a good looking chart for better visualization is time consuming. There are plenty of well-designed charts on the Web, which are ideal references for imitation of chart style. However, stored as bitmap images, reference charts have hinder machine interpretation of style settings and thus difficult to be directly applied. In this paper, we extract visual properties from reference chart images as style templates to restyle charts. We first construct a large-scale dataset of 187,059 chart images from real world data, labeled with predefined visual property values. Then we introduce an end-to-end learning network to extract the properties based on two image-encoding approaches. Furthermore, in order to capture spatial relationships of chart objects, which are crucial in solving the task, we propose a novel positional encoding method to integrate clues of relative positions between objects. Experimental results show that our model significantly outperforms baseline models. By adding positional features, our model achieves better performance. Finally, we present the application for chart restyling based on our model.

Coded aperture snapshot spectral imaging (CASSI) captures a full frame spectral image as a single compressive image and is mandatory to reconstruct the underlying hyperspectral image (HSI) from the snapshot as the post-processing, which is challenge inverse problem due to its ill-posed nature. Existing methods for HSI reconstruction from a snapshot usually employs optimization for solving the formulated image degradation model regularized with the empirically designed priors, and still cannot achieve enough reconstruction accuracy for real HS image analysis systems. Motivated by the recent advances of deep learning for different inverse problems, deep learning based HSI reconstruction method has attracted a lot of attention, and can boost the reconstruction performance. This study proposes a novel deep convolutional neural network (DCNN) based framework for effectively learning the spatial structure and spectral attribute in the underlying HSI with the reciprocal spatial and spectral modules. Further, to adaptively leverage the useful learned feature for better HSI image reconstruction, we integrate residual attention modules into our DCNN via exploring both spatial and spectral attention maps. Experimental results on two benchmark HSI datasets show that our method outperforms state-of-the-art methods in both quantitative values and visual effect.

In this paper we propose a novel easily reproducible technique to attack the best public Face ID system ArcFace in different shooting conditions. To create an attack, we print the rectangular paper sticker on a common color printer and put it on the hat. The adversarial sticker is prepared with a novel algorithm for off-plane transformations of the image which imitates sticker location on the hat. Such an approach confuses the state-of-the-art public Face ID model LResNet100E-IR, ArcFace@ms1m-refine-v2 and is transferable to other Face ID models.

In this paper we consider a collection of relative pose problems which arise naturally in applications for visual indoor UAV navigation. We focus on cases where additional information from an onboard IMU is available and thus provides a partial extrinsic calibration through the gravitational vector. The solvers are designed for a partially calibrated camera, for a variety of realistic indoor scenarios, which makes it possible to navigate using images of the ground floor. Current state-of-the-art solvers use more general assumptions, such as using arbitrary planar structures; however, these solvers do not yield adequate reconstructions for real scenes, nor do they perform fast enough to be incorporated in real-time systems. We show that the proposed solvers enjoy better numerical stability, are faster, and require fewer point correspondences, compared to state-of-the-art solvers. These properties are vital components for robust navigation in real-time systems, and we demonstrate on both synthetic and real data that our method outperforms other methods, and yields superior motion estimation.

Medical image segmentation has played an important role in medical analysis and widely developed for many clinical applications. Deep learning-based approaches have achieved high performance in semantic segmentation but they are limited to pixel-wise setting and imbalanced classes data problem. In this paper, we tackle those limitations by developing a new deep learning-based model which takes into account both higher feature level i.e. region inside contour, intermediate feature level i.e. offset curves around the contour and lower feature level i.e. contour. Our proposed Offset Curves (OsC) loss consists of three main fitting terms. The first fitting term focuses on pixel-wise level segmentation whereas the second fitting term acts as attention model which pays attention to the area around the boundaries (offset curves). The third terms plays a role as regularization term which takes the length of boundaries into account. We evaluate our proposed OsC loss on both 2D network and 3D network. Two common medical datasets, i.e. retina DRIVE and brain tumor BRATS 2018 datasets are used to benchmark our proposed loss performance. The experiments have showed that our proposed OsC loss function outperforms other mainstream loss functions such as Cross-Entropy, Dice, Focal on the most common segmentation networks Unet, FCN.

Existing networks for stereo matching usually use 2-D CNN as the feature extractor. However, objects are usually continuous in spatial, if an extracted feature contains disparity edge (the representation of this feature on original image contains disparity edge), then this feature usually not occur inside the region of an object. We propose a novel attention mechanism to suppress features containing disparity edge, named SDE-Attention (SDEA). We notice that features containing disparity edge are usually continuous in one image and discontinuous in another, which means that they usually have a greater difference in two feature maps of the same layer than features that don’t contain disparity edge. SDEA calculate the weight matrix of the intermediate feature map according to this trait, then the weight matrix is multiplied to the intermediate feature map. We test SDEA on PSMNet, experimental results show that our method has a significant improvement in accuracy and our network achieves state-of-the-art performance among the published networks.

In this paper we investigate the exploitation of non linear scaling of distances for advanced nearest neighbor classification. Starting from the recently found relation between the Hypersphere Classifier (HC) and the Adaptive Nearest Neighbor rule (ANN), here we propose PowerHC, an improved version of HC in which distances are normalized using a non linear mapping; non linear scaling of data, whose usefulness for feature spaces has been already assessed, has been hardly investigated for distances. A thorough experimental evaluation, involving 24 datasets and a challenging real world scenario of seismic signal classification, confirms the suitability of the proposed approach.

Video moment retrieval aims to search the moment most relevant to a given language query. However, most existing methods in this community often require temporal boundary annotations which are expensive and time-consuming to label. Hence weakly supervised methods have been put forward recently by only using coarse video-level label. Despite effectiveness, these methods usually process moment candidates independently, while ignoring a critical issue that the natural temporal dependencies between candidates in different temporal scales. To cope with this issue, we propose a Multi-scale 2D Representation Learning method for weakly supervised video moment retrieval. Specifically, we first construct a two-dimensional map for each temporal scale to capture the temporal dependencies between candidates. Two dimensions in this map indicate the start and end time points of these candidates. Then, we select top-K candidates from each scale-varied map with a learnable convolutional neural network. With a newly designed Moments Evaluation Module, we obtain the alignment scores of the selected candidates. At last, the similarity between captions and language query is served as supervision for further training the candidates' selector. Experiments on two benchmark datasets Charades-STA and ActivityNet Captions demonstrate that our approach achieves superior performance to state-of-the-art results.

Atrial fibrillation is the most common heart rhythm disease. Due to a lack of understanding in matter of underlying atrial structures, current treatments are still not satisfying. Recently, with the popularity of deep learning, many segmentation methods based on fully convolutional networks have been proposed to analyze atrial structures, especially from late gadolinium-enhanced magnetic resonance imaging. However, two problems still occur: 1) segmentation results include the atrial-like background; 2) boundaries are very hard to segment. Most segmentation approaches design a specific network that mainly focuses on the regions, to the detriment of the boundaries. Therefore, this paper proposes an attention full convolutional network framework based on the ResNet-101 architecture, which focuses on boundaries as much as on regions. The additional attention module is added to have the network pay more attention on regions and then to reduce the impact of the misleading similarity of neighboring tissues. We also use a hybrid loss composed of a region loss and a boundary loss to treat boundaries and regions at the same time. We demonstrate the efficiency of the proposed approach on the MICCAI 2018 Atrial Segmentation Challenge public dataset.

We present a novel approach to adjust global image properties such as colour, saturation, and luminance using human-interpretable image enhancement curves, inspired by the Photoshop curves tool. Our method, dubbed neural CURve Layers (CURL), is designed as a multi-colour space neural retouching block trained jointly in three different colour spaces (HSV, CIELab, RGB) guided by a novel multi-colour space loss. The curves are fully differentiable and are trained end-to-end for different computer vision problems including photo enhancement (RGB-to-RGB) and as part of the image signal processing pipeline for image formation (RAW-to-RGB). To demonstrate the effectiveness of CURL we combine this global image transformation block with a pixel-level (local) image multi-scale encoder-decoder backbone network. In an extensive experimental evaluation we show that CURL produces state-of-the-art image quality versus recently proposed deep learning approaches in both objective and perceptual metrics, setting new state-of-the-art performance on multiple public datasets.

Knowledge distillation (KD) is commonly deemed as an effective model compression technique in which a compact model (student) is trained under the supervision of a larger pretrained model or an ensemble of models (teacher). Various techniques have been proposed since the original formulation, which mimics different aspects of the teacher such as the representation space, decision boundary or intra-data relationship. Some methods replace the one way knowledge distillation from a static teacher with collaborative learning between a cohort of students. Despite the recent advances, a clear understanding of where knowledge resides in a deep neural network and optimal method for capturing knowledge from teacher and transferring it to student still remains an open question. In this study we provide an extensive study on 9 different knowledge distillation methods which covers a broad spectrum of approaches to capture and transfer knowledge. We demonstrate the versatility of the KD framework on different datasets and network architectures under varying capacity gaps between the teacher and student. The study provides intuition for the effects of mimicking different aspects of the teacher and derives insights from the performance of the different distillation approaches to guide the the design of more effective KD methods . Furthermore, our study shows the effectiveness of the KD framework in learning efficiently under varying severity levels of label noise and class imbalance, consistently providing significant generalization gains over standard training. We emphasize that the efficacy of KD goes much beyond a model compression technique and should be considered as a general purpose training paradigm which offers more robustness to common challenges in the real-world datasets compared to the standard training procedure.

Point clouds obtained from real word scanning are always incomplete and ununiformly distributed, which would cause structural losses in 3D shape representations. Therefore, a learning-based method is introduced in this paper to repair partial point clouds and restore the complete shapes of target objects. First, we design an encoder that takes both local features and global features into consideration. Second, we establish a graph to connect the local features together, and then implement graph convolution with multi-head attention on it. The graph enables each local feature vector to search across the regions and selectively absorb other local features based on the its own features and global features. Third, we design a coarse decoder to collect cross-region features from the graph and generate coarse point clouds with low resolution, and a folding-based decoder to generate fine point clouds with high resolution. Our network is trained on six categories of objects in the ModelNet dataset, and its performance is compared with several existing methods, the results show that our network is able to generate dense complete point cloud with the highest accuracy.

Aiming at recognizing images of the same person across distinct camera views, person re-identification (re-ID) has been among active research topics in computer vision. Most existing re-ID works require collection of a large amount of labeled image data from the scenes of interest. When the data to be recognized are different from the source-domain training ones, a number of domain adaptation approaches have been proposed. Nevertheless, one still needs to collect labeled or unlabelled target-domain data during training. In this paper, we tackle an even more challenging and practical setting, domain generalized (DG) person re-ID. That is, while a number of labeled source-domain datasets are available, we do not have access to any target-domain training data. In order to learn domain-invariant features without knowing the target domain of interest, we present an episodic learning scheme which advances meta learning strategies to exploit the observed source-domain labeled data. The learned features would exhibit sufficient domain-invariant properties while not overfitting the source-domain data or ID labels. Our experiments on four benchmark datasets confirm the superiority of our method over the state-of-the-arts.

In this paper, we propose an occlusion-robust pose estimation method of an unknown object instance in an object category from a depth image. In a cluttered environment, objects are often occluded mutually. For estimating the pose of an object in such a situation, a method that de-occludes the unobservable area of the object would be effective. However, there are two difficulties; occlusion causes offset between the actual object center and the center of its observable area, and different instances in a category may have different shapes. To cope with these difficulties, we propose a two-stage Encoder-Decoder model to extract features with objects whose centers are aligned to the image center. In the model, we also propose the Median-shape Reconstructor as the second stage to absorb shape variations in a category. By evaluating the method with both a large-scale virtual dataset and a real dataset, we confirmed the proposed method achieves good performance on pose estimation of an occluded object from a depth image.

We propose a CNN-based approach to classify ten genres of ballroom dances given audio recordings, five latin and five standard, namely Cha Cha Cha, Jive, Paso Doble, Rumba, Samba, Quickstep, Slow Foxtrot, Slow Waltz, Tango and Viennese Waltz. We utilize a spectrogram of an audio signal and we treat it as an image that is an input of the CNN. The classification is performed independently by 5-seconds spectrogram segments in sliding window fashion and the results are then aggregated. The method was tested on following datasets: Publicly available Extended Ballroom dataset collected by Marchand and Peeters, 2016 and two YouTube datasets collected by us, one in studio quality and the other, more challenging, recorded on mobile phones. The method achieved accuracy 93.9%, 96.7% and 89.8% respectively. The method runs in real-time. We implemented a web application to demonstrate the proposed method.