Search

In this paper, we propose a novel deep Efficient Relational Sentence Ordering Network (referred to as ERSON) by leveraging pre-trained language model in both encoder and decoder architectures to strengthen the coherence modeling of the entire model. Specifically, we first introduce a divide-and-fuse BERT (referred to as DF-BERT), a new refactor of BERT network, where lower layers in the improved model encode each sentence in the paragraph independently, which are shared by different sentence pairs, and the higher layers learn the cross-attention between sentence pairs jointly. It enables us to capture the semantic concepts and contextual information between the sentences of the paragraph, while significantly reducing the runtime and memory consumption without sacrificing the model performance. Besides, a Relational Pointer Decoder (referred to as RPD) is developed, which utilizes the pre-trained Next Sentence Prediction (NSP) task of BERT to capture the useful relative ordering information between sentences to enhance the order predictions. In addition, a variety of knowledge distillation based losses are added as auxiliary supervision to further improve the ordering performance. The extensive evaluations on Sentence Ordering, Order Discrimination, and Multi-Document Summarization tasks show the superiority of ERSON to the state-of-the-art ordering methods. [ABSTRACT FROM AUTHOR]

Dynamic routing networks, aimed at finding the best routing paths in the networks, have achieved significant improvements to neural networks in terms of accuracy and efficiency. In this paper, we see dynamic routing networks in a fresh light, formulating a routing method as a mapping from a sample space to a routing space. From the perspective of space mapping, prevalent methods of dynamic routing did not take into account how routing paths would be distributed in the routing space. Thus, we propose a novel method, termed CoDiNet, to model the relationship between a sample space and a routing space by regularizing the distribution of routing paths with the properties of consistency and diversity. In principle, the routing paths for the self-supervised similar samples should be closely distributed in the routing space.Moreover, we design a customizable dynamic routing module, which can strike a balance between accuracy and efficiency. When deployed upon ResNet models, our method achieves higher performance and effectively reduces average computational cost on four widely used datasets. [ABSTRACT FROM AUTHOR]

The foundational assumption of machine learning is that the data under consideration is separable into classes; while intuitively reasonable, separability constraints have proven remarkably difficult to formulate mathematically. We believe this problem is rooted in the mismatch between existing statistical techniques and commonly encountered data; object representations are typically high dimensional but statistical techniques tend to treat high dimensions a degenerate case. To address this problem, we develop a dedicated statistical framework for machine learning in high dimensions. The framework derives from the observation that object relations form a natural hierarchy; this leads us to model objects as instances of a high dimensional, hierarchal generative processes. Using a distance based statistical technique, also developed in this paper, we show that in such generative processes, instances of each process in the hierarchy, are almost-always encapsulated by a distinctive-shell that excludes almost-all other instances. The result is shell theory, a statistical machine learning framework in which separability constraints (distinctive-shells) are formally derived from the assumed generative process. [ABSTRACT FROM AUTHOR]

A lot of work has been done towards reconstructing the 3D facial structure from single images by capitalizing on the power of deep convolutional neural networks (DCNNs). In the recent works, the texture features either correspond to components of a linear texture space or are learned by auto-encoders directly from in-the-wild images. In all cases, the quality of the facial texture reconstruction is still not capable of modeling facial texture with high-frequency details. In this paper, we take a radically different approach and harness the power of generative adversarial networks (GANs) and DCNNs in order to reconstruct the facial texture and shape from single images. That is, we utilize GANs to train a very powerful facial texture prior from a large-scale 3D texture dataset. Then, we revisit the original 3D Morphable Models (3DMMs) fitting making use of non-linear optimization to find the optimal latent parameters that best reconstruct the test image but under a new perspective. In order to be robust towards initialisation and expedite the fitting process, we propose a novel self-supervised regression based approach. We demonstrate excellent results in photorealistic and identity preserving 3D face reconstructions and achieve for the first time, to the best of our knowledge, facial texture reconstruction with high-frequency details. [ABSTRACT FROM AUTHOR]

Sparsity-constrained optimization problems are common in machine learning, such as sparse coding, low-rank minimization and compressive sensing. However, most of previous studies focused on constructing various hand-crafted sparse regularizers, while little work was devoted to learning adaptive sparse regularizers from given input data for specific tasks. In this paper, we propose a deep sparse regularizer learning model that learns data-driven sparse regularizers adaptively. Via the proximal gradient algorithm, we find that the sparse regularizer learning is equivalent to learning a parameterized activation function. This encourages us to learn sparse regularizers in the deep learning framework. Therefore, we build a neural network composed of multiple blocks, each being differentiable and reusable. All blocks contain learnable piecewise linear activation functions which correspond to the sparse regularizer to be learned. Furthermore, the proposed model is trained with back propagation, and all parameters in this model are learned end-to-end. We apply our framework to multi-view clustering and semi-supervised classification tasks to learn a latent compact representation. Experimental results demonstrate the superiority of the proposed framework over state-of-the-art multi-view learning models. [ABSTRACT FROM AUTHOR]

We focus on a fundamental task of detecting meaningful line structures, a.k.a., semantic line, in natural scenes. Many previous methods regard this problem as a special case of object detection and adjust existing object detectors for semantic line detection. However, these methods neglect the inherent characteristics of lines, leading to sub-optimal performance. Lines enjoy much simpler geometric property than complex objects and thus can be compactly parameterized by a few arguments. To better exploit the property of lines, in this paper, we incorporate the classical Hough transform technique into deeply learned representations and propose a one-shot end-to-end learning framework for line detection. By parameterizing lines with slopes and biases, we perform Hough transform to translate deep representations into the parametric domain, in which we perform line detection. Specifically, we aggregate features along candidate lines on the feature map plane and then assign the aggregated features to corresponding locations in the parametric domain. Consequently, the problem of detecting semantic lines in the spatial domain is transformed into spotting individual points in the parametric domain, making the post-processing steps, i.e., non-maximal suppression, more efficient. Furthermore, our method makes it easy to extract contextual line features that are critical for accurate line detection. In addition to the proposed method, we design an evaluation metric to assess the quality of line detection and construct a large scale dataset for the line detection task. Experimental results on our proposed dataset and another public dataset demonstrate the advantages of our method over previous state-of-the-art alternatives. The dataset and source code is available at https://mmcheng.net/dhtline/. [ABSTRACT FROM AUTHOR]

This paper investigates the principles of embedding learning to tackle the challenging semi-supervised video object segmentation. Unlike previous practices that focus on exploring the embedding learning of foreground object (s), we consider background should be equally treated. Thus, we propose a Collaborative video object segmentation by Foreground-Background Integration (CFBI) approach. CFBI separates the feature embedding into the foreground object region and its corresponding background region, implicitly promoting them to be more contrastive and improving the segmentation results accordingly. Moreover, CFBI performs both pixel-level matching processes and instance-level attention mechanisms between the reference and the predicted sequence, making CFBI robust to various object scales. Based on CFBI, we introduce a multi-scale matching structure and propose an Atrous Matching strategy, resulting in a more robust and efficient framework, CFBI+. We conduct extensive experiments on two popular benchmarks, i.e., DAVIS and YouTube-VOS. Without applying any simulated data for pre-training, our CFBI+ achieves the performance ($\mathcal {J}$ J & $\mathcal {F}$ F ) of 82.9 and 82.8 percent, outperforming all the other state-of-the-art methods. Code: https://github.com/z-x-yang/CFBI. [ABSTRACT FROM AUTHOR]

Deep learning recognition approaches can potentially perform better if we can extract a discriminative representation that controllably separates nuisance factors. In this paper, we propose a novel approach to explicitly enforce the extracted discriminative representation $\boldsymbol{d}$ d , extracted latent variation $\boldsymbol{l}$ l (e,g., background, unlabeled nuisance attributes), and semantic variation label vector $\boldsymbol{s}$ s (e.g., labeled expressions/pose) to be independent and complementary to each other. We can cast this problem as an adversarial game in the latent space of an auto-encoder. Specifically, with the to-be-disentangled $\boldsymbol{s}$ s , we propose to equip an end-to-end conditional adversarial network with the ability to decompose an input sample into ${\boldsymbol{d}}$ d and $\boldsymbol{l}$ l . However, we argue that maximizing the cross-entropy loss of semantic variation prediction from $\boldsymbol{d}$ d is not sufficient to remove the impact of $\boldsymbol{s}$ s from $\boldsymbol{d}$ d , and that the uniform-target and entropy regularization are necessary. A collaborative mutual information regularization framework is further proposed to avoid unstable adversarial training. It is able to minimize the differentiable mutual information between the variables to enforce independence. The proposed discriminative representation inherits the desired tolerance property guided by prior knowledge of the task. Our proposed framework achieves top performance on diverse recognition tasks, including digits classification, large-scale face recognition on LFW and IJB-A datasets, and face recognition tolerant to changes in lighting, makeup, disguise, etc. [ABSTRACT FROM AUTHOR]

Given a natural language expression and an image/video, the goal of referring segmentation is to produce the pixel-level masks of the entities described by the subject of the expression. Previous approaches tackle this problem by implicit feature interaction and fusion between visual and linguistic modalities in a one-stage manner. However, human tends to solve the referring problem in a progressive manner based on informative words in the expression, i.e., first roughly locating candidate entities and then distinguishing the target one. In this paper, we propose a cross-modal progressive comprehension (CMPC) scheme to effectively mimic human behaviors and implement it as a CMPC-I (Image) module and a CMPC-V (Video) module to improve referring image and video segmentation models. For image data, our CMPC-I module first employs entity and attribute words to perceive all the related entities that might be considered by the expression. Then, the relational words are adopted to highlight the target entity as well as suppress other irrelevant ones by spatial graph reasoning. For video data, our CMPC-V module further exploits action words based on CMPC-I to highlight the correct entity matched with the action cues by temporal graph reasoning. In addition to the CMPC, we also introduce a simple yet effective Text-Guided Feature Exchange (TGFE) module to integrate the reasoned multimodal features corresponding to different levels in the visual backbone under the guidance of textual information. In this way, multi-level features can communicate with each other and be mutually refined based on the textual context. Combining CMPC-I or CMPC-V with TGFE can form our image or video version referring segmentation frameworks and our frameworks achieve new state-of-the-art performances on four referring image segmentation benchmarks and three referring video segmentation benchmarks respectively. Our code is available at https://github.com/spyflying/CMPC-Refseg. [ABSTRACT FROM AUTHOR]

In this paper, we propose a novel system named Disp R-CNN for 3D object detection from stereo images. Many recent works solve this problem by first recovering point clouds with disparity estimation and then apply a 3D detector. The disparity map is computed for the entire image, which is costly and fails to leverage category-specific prior. In contrast, we design an instance disparity estimation network (iDispNet) that predicts disparity only for pixels on objects of interest and learns a category-specific shape prior for more accurate disparity estimation. To address the challenge from scarcity of disparity annotation in training, we propose to use a statistical shape model to generate dense disparity pseudo-ground-truth without the need of LiDAR point clouds, which makes our system more widely applicable. Experiments on the KITTI dataset show that, when LiDAR ground-truth is not used at training time, Disp R-CNN outperforms previous state-of-the-art methods based on stereo input by 20 percent in terms of average precision for all categories. The code and pseudo-ground-truth data are available at the project page: https://github.com/zju3dv/disprcnn. [ABSTRACT FROM AUTHOR]

We tackle human image synthesis, including human motion imitation, appearance transfer, and novel view synthesis, within a unified framework. It means that the model, once being trained, can be used to handle all these tasks. The existing task-specific methods mainly use 2D keypoints (pose) to estimate the human body structure. However, they only express the position information with no ability to characterize the personalized shape of the person and model the limb rotations. In this paper, we propose to use a 3D body mesh recovery module to disentangle the pose and shape. It can not only model the joint location and rotation but also characterize the personalized body shape. To preserve the source information, such as texture, style, color, and face identity, we propose an Attentional Liquid Warping GAN with Attentional Liquid Warping Block (AttLWB) that propagates the source information in both image and feature spaces to the synthesized reference. Specifically, the source features are extracted by a denoising convolutional auto-encoder for characterizing the source identity well. Furthermore, our proposed method can support a more flexible warping from multiple sources. To further improve the generalization ability of the unseen source images, a one/few-shot adversarial learning is applied. In detail, it first trains a model in an extensive training set. Then, it finetunes the model by one/few-shot unseen image(s) in a self-supervised way to generate high-resolution ($512 \times 512$ 512 × 512 and $1024 \times 1024$ 1024 × 1024 ) results. Also, we build a new dataset, namely Impersonator (iPER) dataset, for the evaluation of human motion imitation, appearance transfer, and novel view synthesis. Extensive experiments demonstrate the effectiveness of our methods in terms of preserving face identity, shape consistency, and clothes details. All codes and dataset are available on https://impersonator.org/work/impersonator-plus-plus.html. [ABSTRACT FROM AUTHOR]

Graph matching involves combinatorial optimization based on edge-to-edge affinity matrix, which can be generally formulated as Lawler's quadratic assignment problem (QAP). This paper presents a QAP network directly learning with the affinity matrix (equivalently the association graph) whereby the matching problem is translated into a constrained vertex classification task. The association graph is learned by an embedding network for vertex classification, followed by Sinkhorn normalization and a cross-entropy loss for end-to-end learning. We further improve the embedding model on association graph by introducing Sinkhorn based matching-aware constraint, as well as dummy nodes to deal with unequal sizes of graphs. To our best knowledge, this is one of the first network to directly learn with the general Lawler's QAP. In contrast, recent deep matching methods focus on the learning of node/edge features in two graphs respectively. We also show how to extend our network to hypergraph matching, and matching of multiple graphs. Experimental results on both synthetic graphs and real-world images show its effectiveness. For pure QAP tasks on synthetic data and QAPLIB benchmark, our method can perform competitively and even surpass state-of-the-art graph matching and QAP solvers with notable less time cost. We provide a project homepage at http://thinklab.sjtu.edu.cn/project/NGM/index.html. [ABSTRACT FROM AUTHOR]

In this paper we present methods for estimating shape from polarisation and shading information, i.e. photo-polarimetric shape estimation, under varying, but unknown, illumination, i.e. in an uncalibrated scenario. We propose several alternative photo-polarimetric constraints that depend upon the partial derivatives of the surface and show how to express them in a unified system of partial differential equations of which previous work is a special case. By careful combination and manipulation of the constraints, we show how to eliminate non-linearities such that a discrete version of the problem can be solved using linear least squares. We derive a minimal, combinatorial approach for two source illumination estimation which we use with RANSAC for robust light direction and intensity estimation. We also introduce a new method for estimating a polarisation image from multichannel data and provide methods for estimating albedo and refractive index. We evaluate lighting, shape, albedo and refractive index estimation methods on both synthetic and real-world data showing improvements over existing state-of-the-art. [ABSTRACT FROM AUTHOR]

Spatially-adaptive normalization (SPADE) is remarkably successful recently in conditional semantic image synthesis in T. Park et al. 2019 which modulates the normalized activation with spatially-varying transformations learned from semantic layouts, to prevent the semantic information from being washed away. Despite its impressive performance, a more thorough understanding of the advantages inside the box is still highly demanded to help reduce the significant computation and parameter overhead introduced by this novel structure. In this paper, from a return-on-investment point of view, we conduct an in-depth analysis of the effectiveness of this spatially-adaptive normalization and observe that its modulation parameters benefit more from semantic-awareness rather than spatial-adaptiveness, especially for high-resolution input masks. Inspired by this observation, we propose class-adaptive normalization (CLADE), a lightweight but equally-effective variant that is only adaptive to semantic class. In order to further improve spatial-adaptiveness, we introduce intra-class positional map encoding calculated from semantic layouts to modulate the normalization parameters of CLADE and propose a truly spatially-adaptive variant of CLADE, namely CLADE-ICPE. Through extensive experiments on multiple challenging datasets, we demonstrate that the proposed CLADE can be generalized to different SPADE-based methods while achieving comparable generation quality compared to SPADE, but it is much more efficient with fewer extra parameters and lower computational cost. The code and pretrained models are available at https://github.com/tzt101/CLADE.git. [ABSTRACT FROM AUTHOR]

Contextual information plays an important role in solving various image and scene understanding tasks. Prior works have focused on the extraction of contextual information from an image and use it to infer the properties of some object(s) in the image or understand the scene behind the image, e.g., context-based object detection, recognition and semantic segmentation. In this paper, we consider an inverse problem, i.e., how to hallucinate the missing contextual information from the properties of standalone objects. We refer to it as object-level scene context prediction. This problem is difficult, as it requires extensive knowledge of the complex and diverse relationships among objects in the scene. We propose a deep neural network, which takes as input the properties (i.e., category, shape, and position) of a few standalone objects to predict an object-level scene layout that compactly encodes the semantics and structure of the scene context where the given objects are. Quantitative experiments and user studies demonstrate that our model can generate more plausible scene contexts than the baselines. Our model also enables the synthesis of realistic scene images from partial scene layouts. Finally, we validate that our model internally learns useful features for scene recognition and fake scene detection. [ABSTRACT FROM AUTHOR]

Systems that are based on recursive Bayesian updates for classification limit the cost of evidence collection through certain stopping/termination criteria and accordingly enforce decision making. Conventionally, two termination criteria based on pre-defined thresholds over (i) the maximum of the state posterior distribution; and (ii) the state posterior uncertainty are commonly used. In this paper, we propose a geometric interpretation over the state posterior progression and accordingly we provide a point-by-point analysis over the disadvantages of using such conventional termination criteria. For example, through the proposed geometric interpretation we show that confidence thresholds defined over maximum of the state posteriors suffer from stiffness that results in unnecessary evidence collection whereas uncertainty based thresholding methods are fragile to number of categories and terminate prematurely if some state candidates are already discovered to be unfavorable. Moreover, both types of termination methods neglect the evolution of posterior updates. We then propose a new stopping/termination criterion with a geometrical insight to overcome the limitations of these conventional methods and provide a comparison in terms of decision accuracy and speed. We validate our claims using simulations and using real experimental data obtained through a brain computer interfaced typing system. [ABSTRACT FROM AUTHOR]

As a challenging task of high-level video understanding, weakly supervised temporal action localization has attracted more attention recently. With only video-level category labels, this task should indistinguishably identify the background and action categories frame by frame. However, it is non-trivial to achieve this in untrimmed videos, due to the unconstrained background, complex and multi-label actions. With the observation that these difficulties are mainly brought by the large variations within background and actions, we propose to address these challenges from the perspective of modeling variations. Moreover, it is desired to further reduce the variations, or learn compact features, so as to cast the problem of background identification as rejecting background and alleviate the contradiction between classification and detection. Accordingly, in this paper, we propose a two-branch relational prototypical network. The first branch, namely action-branch, adopts class-wise prototypes and mainly acts as an auxiliary to introduce priori knowledge about label dependencies and be a guide for the second branch. Meanwhile, the second branch, namely sub-branch, starts with multiple prototypes, namely sub-prototypes, to enable a powerful ability of modeling variations. As a further benefit, we elaborately design a multi-label clustering loss based on the sub-prototypes to learn compact features under the multi-label setting. The two branches are associated using the correspondences between two types of prototypes, leading to a special two-stage classifier in the s-branch, on the other hand, the two branches serve as regularization terms to each other, improving the final performance. Ablation studies find that the proposed model is capable of modeling classes with large variations and learning compact features. Extensive experimental evaluations on Thumos14, MultiThumos and ActivityNet datasets demonstrate the effectiveness of the proposed method and superior performance over state-of-the-art approaches. [ABSTRACT FROM AUTHOR]

Kernel-based methods for support vector machines (SVM) have shown highly advantageous performance in various applications. However, they may incur prohibitive computational costs for large-scale sample datasets. Therefore, data reduction (reducing the number of support vectors) appears to be necessary, which gives rise to the topic of the sparse SVM. Motivated by this problem, the sparsity constrained kernel SVM optimization has been considered in this paper in order to control the number of support vectors. Based on the established optimality conditions associated with the stationary equations, a Newton-type method is developed to handle the sparsity constrained optimization. This method is found to enjoy the one-step convergence property if the starting point is chosen to be close to a local region of a stationary point, thereby leading to a super-high computational speed. Numerical comparisons with several powerful solvers demonstrate that the proposed method performs exceptionally well, particularly for large-scale datasets in terms of a much lower number of support vectors and shorter computational time. [ABSTRACT FROM AUTHOR]

Existing RGB-D salient object detection (SOD) models usually treat RGB and depth as independent information and design separate networks for feature extraction from each. Such schemes can easily be constrained by a limited amount of training data or over-reliance on an elaborately designed training process. Inspired by the observation that RGB and depth modalities actually present certain commonality in distinguishing salient objects, a novel joint learning and densely cooperative fusion (JL-DCF) architecture is designed to learn from both RGB and depth inputs through a shared network backbone, known as the Siamese architecture. In this paper, we propose two effective components: joint learning (JL), and densely cooperative fusion (DCF). The JL module provides robust saliency feature learning by exploiting cross-modal commonality via a Siamese network, while the DCF module is introduced for complementary feature discovery. Comprehensive experiments using five popular metrics show that the designed framework yields a robust RGB-D saliency detector with good generalization. As a result, JL-DCF significantly advances the state-of-the-art models by an average of $\sim 2.0\%$ ∼ 2. 0 % (max F-measure) across seven challenging datasets. In addition, we show that JL-DCF is readily applicable to other related multi-modal detection tasks, including RGB-T (thermal infrared) SOD and video SOD, achieving comparable or even better performance against state-of-the-art methods. We also link JL-DCF to the RGB-D semantic segmentation field, showing its capability of outperforming several semantic segmentation models on the task of RGB-D SOD. These facts further confirm that the proposed framework could offer a potential solution for various applications and provide more insight into the cross-modal complementarity task. [ABSTRACT FROM AUTHOR]

The light field (LF) reconstruction is mainly confronted with two challenges, large disparity and the non-Lambertian effect. Typical approaches either address the large disparity challenge using depth estimation followed by view synthesis or eschew explicit depth information to enable non-Lambertian rendering, but rarely solve both challenges in a unified framework. In this paper, we revisit the classic LF rendering framework to address both challenges by incorporating it with advanced deep learning techniques. First, we analytically show that the essential issue behind the large disparity and non-Lambertian challenges is the aliasing problem. Classic LF rendering approaches typically mitigate the aliasing with a reconstruction filter in the Fourier domain, which is, however, intractable to implement within a deep learning pipeline. Instead, we introduce an alternative framework to perform anti-aliasing reconstruction in the image domain and analytically show comparable efficacy on the aliasing issue. To explore the full potential, we then embed the anti-aliasing framework into a deep neural network through the design of an integrated architecture and trainable parameters. The network is trained through end-to-end optimization using a peculiar training set, including regular LFs and unstructured LFs. The proposed deep learning pipeline shows a substantial superiority in solving both the large disparity and the non-Lambertian challenges compared with other state-of-the-art approaches. In addition to the view interpolation for an LF, we also show that the proposed pipeline also benefits light field view extrapolation. [ABSTRACT FROM AUTHOR]

In this paper, we reveal that metric learning would suffer from serious inseparable problem if without informative sample mining. Since the inseparable samples are often mixed with hard samples, current informative sample mining strategies used to deal with inseparable problem may bring up some side-effects, such as instability of objective function, etc. To alleviate this problem, we propose a novel distance metric learning algorithm, named adaptive neighborhood metric learning (ANML). In ANML, we design two thresholds to adaptively identify the inseparable similar and dissimilar samples in the training procedure, thus inseparable sample removing and metric parameter learning are implemented in the same procedure. Due to the non-continuity of the proposed ANML, we develop an ingenious function, named log-exp mean function to construct a continuous formulation to surrogate it, which can be efficiently solved by the gradient descent method. Similar to Triplet loss, ANML can be used to learn both the linear and deep embeddings. By analyzing the proposed method, we find it has some interesting properties. For example, when ANML is used to learn the linear embedding, current famous metric learning algorithms such as the large margin nearest neighbor (LMNN) and neighbourhood components analysis (NCA) are the special cases of the proposed ANML by setting the parameters different values. When it is used to learn deep features, the state-of-the-art deep metric learning algorithms such as Triplet loss, Lifted structure loss, and Multi-similarity loss become the special cases of ANML. Furthermore, the log-exp mean function proposed in our method gives a new perspective to review the deep metric learning methods such as Prox-NCA and N-pairs loss. At last, promising experimental results demonstrate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Representations in the form of Symmetric Positive Definite (SPD) matrices have been popularized in a variety of visual learning applications due to their demonstrated ability to capture rich second-order statistics of visual data. There exist several similarity measures for comparing SPD matrices with documented benefits. However, selecting an appropriate measure for a given problem remains a challenge and in most cases, is the result of a trial-and-error process. In this paper, we propose to learn similarity measures in a data-driven manner. To this end, we capitalize on the $\alpha \beta$ α β -log-det divergence, which is a meta-divergence parametrized by scalars $\alpha$ α and $\beta$ β , subsuming a wide family of popular information divergences on SPD matrices for distinct and discrete values of these parameters. Our key idea is to cast these parameters in a continuum and learn them from data. We systematically extend this idea to learn vector-valued parameters, thereby increasing the expressiveness of the underlying non-linear measure. We conjoin the divergence learning problem with several standard tasks in machine learning, including supervised discriminative dictionary learning and unsupervised SPD matrix clustering. We present Riemannian gradient descent schemes for optimizing our formulations efficiently, and show the usefulness of our method on eight standard computer vision tasks. [ABSTRACT FROM AUTHOR]

This paper generalizes the Attention in Attention (AiA) mechanism, in P. Fang et al., 2019 by employing explicit mapping in reproducing kernel Hilbert spaces to generate attention values of the input feature map. The AiA mechanism models the capacity of building inter-dependencies among the local and global features by the interaction of inner and outer attention modules. Besides a vanilla AiA module, termed linear attention with AiA, two non-linear counterparts, namely, second-order polynomial attention and Gaussian attention, are also proposed to utilize the non-linear properties of the input features explicitly, via the second-order polynomial kernel and Gaussian kernel approximation. The deep convolutional neural network, equipped with the proposed AiA blocks, is referred to as Attention in Attention Network (AiA-Net). The AiA-Net learns to extract a discriminative pedestrian representation, which combines complementary person appearance and corresponding part features. Extensive ablation studies verify the effectiveness of the AiA mechanism and the use of non-linear features hidden in the feature map for attention design. Furthermore, our approach outperforms current state-of-the-art by a considerable margin across a number of benchmarks. In addition, state-of-the-art performance is also achieved in the video person retrieval task with the assistance of the proposed AiA blocks. [ABSTRACT FROM AUTHOR]

Pixel-level 2D object semantic understanding is an important topic in computer vision and could help machine deeply understand objects (e.g., functionality and affordance) in our daily life. However, most previous methods directly train on correspondences in 2D images, which is end-to-end but loses plenty of information in 3D spaces. In this paper, we propose a new method on predicting image corresponding semantics in 3D domain and then projecting them back onto 2D images to achieve pixel-level understanding. In order to obtain reliable 3D semantic labels that are absent in current image datasets, we build a large scale keypoint knowledge engine called KeypointNet, which contains 103,450 keypoints and 8,234 3D models from 16 object categories. Our method leverages the advantages in 3D vision and can explicitly reason about objects self-occlusion and visibility. We show that our method gives comparative and even superior results on standard semantic benchmarks. [ABSTRACT FROM AUTHOR]

This paper explores the non-convex composition optimization consisting of inner and outer finite-sum functions with a large number of component functions. This problem arises in important applications such as nonlinear embedding and reinforcement learning. Although existing approaches such as stochastic gradient descent (SGD) and stochastic variance reduced gradient (SVRG) descent can be applied to solve this problem, their query complexities tend to be high, especially when the number of inner component functions is large. Therefore, to significantly improve the query complexity of current approaches, we have devised the stochastic composition via variance reduction (SCVR). What's more, we analyze the query complexity under different numbers of inner function and outer function. Based on different kinds of estimation of inner component function, we also present the SCVRII algorithm, though the order of query complexities are the same with SCVR. Additionally, we propose an extension to handle the mini-batch cases, which improve the query complexity under the optimal mini-batch size. The experimental results validate our proposed algorithms and theoretical analyses. [ABSTRACT FROM AUTHOR]

Stereo matching is one of the most popular techniques to estimate dense depth maps by finding the disparity between matching pixels on two, synchronized and rectified images. Alongside with the development of more accurate algorithms, the research community focused on finding good strategies to estimate the reliability, i.e., the confidence, of estimated disparity maps. This information proves to be a powerful cue to naively find wrong matches as well as to improve the overall effectiveness of a variety of stereo algorithms according to different strategies. In this paper, we review more than ten years of developments in the field of confidence estimation for stereo matching. We extensively discuss and evaluate existing confidence measures and their variants, from hand-crafted ones to the most recent, state-of-the-art learning based methods. We study the different behaviors of each measure when applied to a pool of different stereo algorithms and, for the first time in literature, when paired with a state-of-the-art deep stereo network. Our experiments, carried out on five different standard datasets, provide a comprehensive overview of the field, highlighting in particular both strengths and limitations of learning-based strategies. [ABSTRACT FROM AUTHOR]

Neural Architecture Search (NAS) has emerged as a promising technique for automatic neural network design. However, existing MCTS based NAS approaches often utilize manually designed action space, which is not directly related to the performance metric to be optimized (e.g., accuracy), leading to sample-inefficient explorations of architectures. To improve the sample efficiency, this paper proposes Latent Action Neural Architecture Search (LaNAS), which learns actions to recursively partition the search space into good or bad regions that contain networks with similar performance metrics. During the search phase, as different action sequences lead to regions with different performance, the search efficiency can be significantly improved by biasing towards the good regions. On three NAS tasks, empirical results demonstrate that LaNAS is at least an order more sample efficient than baseline methods including evolutionary algorithms, Bayesian optimizations, and random search. When applied in practice, both one-shot and regular LaNAS consistently outperform existing results. Particularly, LaNAS achieves 99.0 percent accuracy on CIFAR-10 and 80.8 percent top1 accuracy at 600 MFLOPS on ImageNet in only 800 samples, significantly outperforming AmoebaNet with $33\times$ 33 × fewer samples. Our code is publicly available at https://github.com/facebookresearch/LaMCTS. [ABSTRACT FROM AUTHOR]

Crowded scene surveillance can significantly benefit from combining egocentric-view and its complementary top-view cameras. A typical setting is an egocentric-view camera, e.g., a wearable camera on the ground capturing rich local details, and a top-view camera, e.g., a drone-mounted one from high altitude providing a global picture of the scene. To collaboratively analyze such complementary-view videos, an important task is to associate and track multiple people across views and over time, which is challenging and differs from classical human tracking, since we need to not only track multiple subjects in each video, but also identify the same subjects across the two complementary views. This paper formulates it as a constrained mixed integer programming problem, wherein a major challenge is how to effectively measure subjects similarity over time in each video and across two views. Although appearance and motion consistencies well apply to over-time association, they are not good at connecting two highly different complementary views. To this end, we present a spatial distribution based approach to reliable cross-view subject association. We also build a dataset to benchmark this new challenging task. Extensive experiments verify the effectiveness of our method. [ABSTRACT FROM AUTHOR]

The $L_{1}$ L 1 -regularized logistic regression (L1-LR) is popular for classification problems. To accelerate its training speed for high-dimensional data, techniques named safe screening rules have been proposed recently. They can safely delete the inactive features in data so as to greatly reduce the training cost of L1-LR. The screening power of these rules is determined by their corresponding safe regions, which is also the core technique of safe screening rules. In this paper, we introduce a new safe feature elimination rule (SFER) for L1-LR. Compared to existing safe rules, the safe region of SFER is improved in two aspects: (1) a smaller sphere region is constructed by using the strong convexity of dual L1-LR twice; (2) multiple half-spaces, which correspond to the potential active constraints, are added for further contraction. Both improvements can enhance the screening ability of SFER. As for the complexity of SFER, an iterative filtering framework is given by decomposing the safe region into multiple “domes”. In this way, SFER admits a closed form solution and the identified features will not be scanned repeatedly. Experiments on ten benchmark data sets demonstrate that SFER gives superior performance than existing methods on training efficiency. [ABSTRACT FROM AUTHOR]

Stereo matching is one of the longest-standing problems in computer vision with close to 40 years of studies and research. Throughout the years the paradigm has shifted from local, pixel-level decision to various forms of discrete and continuous optimization to data-driven, learning-based methods. Recently, the rise of machine learning and the rapid proliferation of deep learning enhanced stereo matching with new exciting trends and applications unthinkable until a few years ago. Interestingly, the relationship between these two worlds is two-way. While machine, and especially deep, learning advanced the state-of-the-art in stereo matching, stereo itself enabled new ground-breaking methodologies such as self-supervised monocular depth estimation based on deep networks. In this paper, we review recent research in the field of learning-based depth estimation from single and binocular images highlighting the synergies, the successes achieved so far and the open challenges the community is going to face in the immediate future. [ABSTRACT FROM AUTHOR]

Human adaptability relies crucially on learning and merging knowledge from both supervised and unsupervised tasks: the parents point out few important concepts, but then the children fill in the gaps on their own. This is particularly effective, because supervised learning can never be exhaustive and thus learning autonomously allows to discover invariances and regularities that help to generalize. In this paper we propose to apply a similar approach to the problem of object recognition across domains: our model learns the semantic labels in a supervised fashion, and broadens its understanding of the data by learning from self-supervised signals on the same images. This secondary task helps the network to focus on object shapes, learning concepts like spatial orientation and part correlation, while acting as a regularizer for the classification task over multiple visual domains. Extensive experiments confirm our intuition and show that our multi-task method, combining supervised and self-supervised knowledge, provides competitive results with respect to more complex domain generalization and adaptation solutions. It also proves its potential in the novel and challenging predictive and partial domain adaptation scenarios. [ABSTRACT FROM AUTHOR]

Cameras currently allow access to two image states: (i) a minimally processed linear raw-RGB image state (i.e., raw sensor data); or (ii) a highly-processed nonlinear image state (e.g., sRGB). There are many computer vision tasks that work best with a linear image state, such as image deblurring and image dehazing. Unfortunately, the vast majority of images are saved in the nonlinear image state. Because of this, a number of methods have been proposed to “unprocess” nonlinear images back to a raw-RGB state. However, existing unprocessing methods have a drawback because raw-RGB images are sensor-specific. As a result, it is necessary to know which camera produced the sRGB output and use a method or network tailored for that sensor to properly unprocess it. This paper addresses this limitation by exploiting another camera image state that is not available as an output, but it is available inside the camera pipeline. In particular, cameras apply a colorimetric conversion step to convert the raw-RGB image to a device-independent space based on the CIE XYZ color space before they apply the nonlinear photo-finishing. Leveraging this canonical image state, we propose a deep learning framework, CIE XYZ Net, that can unprocess a nonlinear image back to the canonical CIE XYZ image. This image can then be processed by any low-level computer vision operator and re-rendered back to the nonlinear image. We demonstrate the usefulness of the CIE XYZ Net on several low-level vision tasks and show significant gains that can be obtained by this processing framework. Code and dataset are publicly available at https://github.com/mahmoudnafifi/CIE_XYZ_NET. [ABSTRACT FROM AUTHOR]

While most existing multilabel ranking methods assume the availability of a single objective label ranking for each instance in the training set, this paper deals with a more common case where only subjective inconsistent rankings from multiple rankers are associated with each instance. Two ranking methods are proposed from the perspective of instances and rankers, respectively. The first method, Instance-oriented Preference Distribution Learning (IPDL), is to learn a latent preference distribution for each instance. IPDL generates a common preference distribution that is most compatible to all the personal rankings, and then learns a mapping from the instances to the preference distributions. The second method, Ranker-oriented Preference Distribution Learning (RPDL), is proposed by leveraging interpersonal inconsistency among rankers, to learn a unified model from personal preference distribution models of all rankers. These two methods are applied to natural scene images dataset and 3D facial expression dataset BU_3DFE. Experimental results show that IPDL and RPDL can effectively incorporate the information given by the inconsistent rankers, and perform remarkably better than the compared state-of-the-art multilabel ranking algorithms. [ABSTRACT FROM AUTHOR]

Label-specific features serve as an effective strategy to learn from multi-label data, where a set of features encoding specific characteristics of each label are generated to help induce multi-label classification model. Existing approaches work by taking the two-stage strategy, where the procedure of label-specific feature generation is independent of the follow-up procedure of classification model induction. Intuitively, the performance of resulting classification model may be suboptimal due to the decoupling nature of the two-stage strategy. In this paper, a wrapped learning approach is proposed which aims to jointly perform label-specific feature generation and classification model induction. Specifically, one (kernelized) linear model is learned for each label where label-specific features are simultaneously generated within an embedded feature space via empirical loss minimization and pairwise label correlation regularization. Comparative studies over a total of sixteen benchmark data sets clearly validate the effectiveness of the wrapped strategy in exploiting label-specific features for multi-label classification. [ABSTRACT FROM AUTHOR]

3D human pose and shape estimation from monocular images has been an active research area in computer vision. Existing deep learning methods for this task rely on high-resolution input, which however, is not always available in many scenarios such as video surveillance and sports broadcasting. Two common approaches to deal with low-resolution images are applying super-resolution techniques to the input, which may result in unpleasant artifacts, or simply training one model for each resolution, which is impractical in many realistic applications. To address the above issues, this paper proposes a novel algorithm called RSC-Net, which consists of a Resolution-aware network, a Self-supervision loss, and a Contrastive learning scheme. The proposed method is able to learn 3D body pose and shape across different resolutions with one single model. The self-supervision loss enforces scale-consistency of the output, and the contrastive learning scheme enforces scale-consistency of the deep features. We show that both these new losses provide robustness when learning in a weakly-supervised manner. Moreover, we extend the RSC-Net to handle low-resolution videos and apply it to reconstruct textured 3D pedestrians from low-resolution input. Extensive experiments demonstrate that the RSC-Net can achieve consistently better results than the state-of-the-art methods for challenging low-resolution images. [ABSTRACT FROM AUTHOR]

An effective person re-identification (re-ID) model should learn feature representations that are both discriminative, for distinguishing similar-looking people, and generalisable, for deployment across datasets without any adaptation. In this paper, we develop novel CNN architectures to address both challenges. First, we present a re-ID CNN termed omni-scale network (OSNet) to learn features that not only capture different spatial scales but also encapsulate a synergistic combination of multiple scales, namely omni-scale features. The basic building block consists of multiple convolutional streams, each detecting features at a certain scale. For omni-scale feature learning, a unified aggregation gate is introduced to dynamically fuse multi-scale features with channel-wise weights. OSNet is lightweight as its building blocks comprise factorised convolutions. Second, to improve generalisable feature learning, we introduce instance normalisation (IN) layers into OSNet to cope with cross-dataset discrepancies. Further, to determine the optimal placements of these IN layers in the architecture, we formulate an efficient differentiable architecture search algorithm. Extensive experiments show that, in the conventional same-dataset setting, OSNet achieves state-of-the-art performance, despite being much smaller than existing re-ID models. In the more challenging yet practical cross-dataset setting, OSNet beats most recent unsupervised domain adaptation methods without using any target data. Our code and models are released at https://github.com/KaiyangZhou/deep-person-reid . [ABSTRACT FROM AUTHOR]

It is difficult to solve complex tasks that involve large state spaces and long-term decision processes by reinforcement learning (RL) algorithms. A common and promising method to address this challenge is to compress a large RL problem into a small one. Towards this goal, the compression should be state-temporal and optimality-preserving (i.e., the optimal policy of the compressed problem should correspond to that of the uncompressed problem). In this paper, we propose a reward-restricted geodesic (RRG) metric, which can be learned by a neural network, to perform state-temporal compression in RL. We prove that compression based on the RRG metric is approximately optimality-preserving for the raw RL problem endowed with temporally abstract actions. With this compression, we design an RRG metric-based reinforcement learning (RRG-RL) algorithm to solve complex tasks. Experiments in both discrete (2D Minecraft) and continuous (Doom) environments demonstrated the superiority of our method over existing RL approaches. [ABSTRACT FROM AUTHOR]

In a real life process evolving over time, the relationship between its relevant variables may change. Therefore, it is advantageous to have different inference models for each state of the process. Asymmetric hidden Markov models fulfil this dynamical requirement and provide a framework where the trend of the process can be expressed as a latent variable. In this paper, we modify these recent asymmetric hidden Markov models to have an asymmetric autoregressive component in the case of continuous variables, allowing the model to choose the order of autoregression that maximizes its penalized likelihood for a given training set. Additionally, we show how inference, hidden states decoding and parameter learning must be adapted to fit the proposed model. Finally, we run experiments with synthetic and real data to show the capabilities of this new model. [ABSTRACT FROM AUTHOR]

It is important and challenging to infer stochastic latent semantics for natural language applications. The difficulty in stochastic sequential learning is caused by the posterior collapse in variational inference. The input sequence is disregarded in the estimated latent variables. This paper proposes three components to tackle this difficulty and build the variational sequence autoencoder (VSAE) where sufficient latent information is learned for sophisticated sequence representation. First, the complementary encoders based on a long short-term memory (LSTM) and a pyramid bidirectional LSTM are merged to characterize global and structural dependencies of an input sequence, respectively. Second, a stochastic self attention mechanism is incorporated in a recurrent decoder. The latent information is attended to encourage the interaction between inference and generation in an encoder-decoder training procedure. Third, an autoregressive Gaussian prior of latent variable is used to preserve the information bound. Different variants of VSAE are proposed to mitigate the posterior collapse in sequence modeling. A series of experiments are conducted to demonstrate that the proposed individual and hybrid sequence autoencoders substantially improve the performance for variational sequential learning in language modeling and semantic understanding for document classification and summarization. [ABSTRACT FROM AUTHOR]

Unlike the conventional facial expressions, micro-expressions are involuntary and transient facial expressions capable of revealing the genuine emotions that people attempt to hide. Therefore, they can provide important information in a broad range of applications such as lie detection, criminal detection, etc. Since micro-expressions are transient and of low intensity, however, their detection and recognition is difficult and relies heavily on expert experiences. Due to its intrinsic particularity and complexity, video-based micro-expression analysis is attractive but challenging, and has recently become an active area of research. Although there have been numerous developments in this area, thus far there has been no comprehensive survey that provides researchers with a systematic overview of these developments with a unified evaluation. Accordingly, in this survey paper, we first highlight the key differences between macro- and micro-expressions, then use these differences to guide our research survey of video-based micro-expression analysis in a cascaded structure, encompassing the neuropsychological basis, datasets, features, spotting algorithms, recognition algorithms, applications and evaluation of state-of-the-art approaches. For each aspect, the basic techniques, advanced developments and major challenges are addressed and discussed. Furthermore, after considering the limitations of existing micro-expression datasets, we present and release a new dataset — called micro-and-macro expression warehouse (MMEW) — containing more video samples and more labeled emotion types. We then perform a unified comparison of representative methods on CAS(ME) $^2$ 2 for spotting, and on MMEW and SAMM for recognition, respectively. Finally, some potential future research directions are explored and outlined. [ABSTRACT FROM AUTHOR]

Large-scale and multidimensional spatiotemporal data sets are becoming ubiquitous in many real-world applications such as monitoring urban traffic and air quality. Making predictions on these time series has become a critical challenge due to not only the large-scale and high-dimensional nature but also the considerable amount of missing data. In this paper, we propose a Bayesian temporal factorization (BTF) framework for modeling multidimensional time series—in particular spatiotemporal data—in the presence of missing values. By integrating low-rank matrix/tensor factorization and vector autoregressive (VAR) process into a single probabilistic graphical model, this framework can characterize both global and local consistencies in large-scale time series data. The graphical model allows us to effectively perform probabilistic predictions and produce uncertainty estimates without imputing those missing values. We develop efficient Gibbs sampling algorithms for model inference and model updating for real-time prediction and test the proposed BTF framework on several real-world spatiotemporal data sets for both missing data imputation and multi-step rolling prediction tasks. The numerical experiments demonstrate the superiority of the proposed BTF approaches over existing state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Remarkable achievements have been obtained by deep neural networks in the last several years. However, the breakthrough in neural networks accuracy is always accompanied by explosive growth of computation and parameters, which leads to a severe limitation of model deployment. In this paper, we propose a novel knowledge distillation technique named self-distillation to address this problem. Self-distillation attaches several attention modules and shallow classifiers at different depths of neural networks and distills knowledge from the deepest classifier to the shallower classifiers. Different from the conventional knowledge distillation methods where the knowledge of the teacher model is transferred to another student model, self-distillation can be considered as knowledge transfer in the same model - from the deeper layers to the shallow layers. Moreover, the additional classifiers in self-distillation allow the neural network to work in a dynamic manner, which leads to a much higher acceleration. Experiments demonstrate that self-distillation has consistent and significant effectiveness on various neural networks and datasets. On average, 3.49 and 2.32 percent accuracy boost are observed on CIFAR100 and ImageNet. Besides, experiments show that self-distillation can be combined with other model compression methods, including knowledge distillation, pruning and lightweight model design. [ABSTRACT FROM AUTHOR]

Image compression is one of the most fundamental techniques and commonly used applications in the image and video processing field. Earlier methods built a well-designed pipeline, and efforts were made to improve all modules of the pipeline by handcrafted tuning. Later, tremendous contributions were made, especially when data-driven methods revitalized the domain with their excellent modeling capacities and flexibility in incorporating newly designed modules and constraints. Despite great progress, a systematic benchmark and comprehensive analysis of end-to-end learned image compression methods are lacking. In this paper, we first conduct a comprehensive literature survey of learned image compression methods. The literature is organized based on several aspects to jointly optimize the rate-distortion performance with a neural network, i.e., network architecture, entropy model and rate control. We describe milestones in cutting-edge learned image-compression methods, review a broad range of existing works, and provide insights into their historical development routes. With this survey, the main challenges of image compression methods are revealed, along with opportunities to address the related issues with recent advanced learning methods. This analysis provides an opportunity to take a further step towards higher-efficiency image compression. By introducing a coarse-to-fine hyperprior model for entropy estimation and signal reconstruction, we achieve improved rate-distortion performance, especially on high-resolution images. Extensive benchmark experiments demonstrate the superiority of our model in rate-distortion performance and time complexity on multi-core CPUs and GPUs. [ABSTRACT FROM AUTHOR]

We study network pruning which aims to remove redundant channels/kernels and hence speed up the inference of deep networks. Existing pruning methods either train from scratch with sparsity constraints or minimize the reconstruction error between the feature maps of the pre-trained models and the compressed ones. Both strategies suffer from some limitations: the former kind is computationally expensive and difficult to converge, while the latter kind optimizes the reconstruction error but ignores the discriminative power of channels. In this paper, we propose a simple-yet-effective method called discrimination-aware channel pruning (DCP) to choose the channels that actually contribute to the discriminative power. To this end, we first introduce additional discrimination-aware losses into the network to increase the discriminative power of the intermediate layers. Next, we select the most discriminative channels for each layer by considering the discrimination-aware loss and the reconstruction error, simultaneously. We then formulate channel pruning as a sparsity-inducing optimization problem with a convex objective and propose a greedy algorithm to solve the resultant problem. Note that a channel (3D tensor) often consists of a set of kernels (each with a 2D matrix). Besides the redundancy in channels, some kernels in a channel may also be redundant and fail to contribute to the discriminative power of the network, resulting in kernel level redundancy. To solve this issue, we propose a discrimination-aware kernel pruning (DKP) method to further compress deep networks by removing redundant kernels. To avoid manually determining the pruning rate for each layer, we propose two adaptive stopping conditions to automatically determine the number of selected channels/kernels. The proposed adaptive stopping conditions tend to yield more efficient models with better performance in practice. Extensive experiments on both image classification and face recognition demonstrate the effectiveness of our methods. For example, on ILSVRC-12, the resultant ResNet-50 model with 30 percent reduction of channels even outperforms the baseline model by 0.36 percent in terms of Top-1 accuracy. We also deploy the pruned models on a smartphone (equipped with a Qualcomm Snapdragon 845 processor). The pruned MobileNetV1 and MobileNetV2 achieve 1.93× and 1.42× inference acceleration on the mobile device, respectively, with negligible performance degradation. The source code and the pre-trained models are available at https://github.com/SCUT-AILab/DCP. [ABSTRACT FROM AUTHOR]

This paper studies the problem of learning the conditional distribution of a high-dimensional output given an input, where the output and input may belong to two different domains, e.g., the output is a photo image and the input is a sketch image. We solve this problem by cooperative training of a fast thinking initializer and slow thinking solver. The initializer generates the output directly by a non-linear transformation of the input as well as a noise vector that accounts for latent variability in the output. The slow thinking solver learns an objective function in the form of a conditional energy function, so that the output can be generated by optimizing the objective function, or more rigorously by sampling from the conditional energy-based model. We propose to learn the two models jointly, where the fast thinking initializer serves to initialize the sampling of the slow thinking solver, and the solver refines the initial output by an iterative algorithm. The solver learns from the difference between the refined output and the observed output, while the initializer learns from how the solver refines its initial output. We demonstrate the effectiveness of the proposed method on various conditional learning tasks, e.g., class-to-image generation, image-to-image translation, and image recovery. The advantage of our method over GAN-based methods is that our method is equipped with a slow thinking process that refines the solution guided by a learned objective function. [ABSTRACT FROM AUTHOR]

This paper proposes a new generative adversarial network for pose transfer, i.e., transferring the pose of a given person to a target pose. We design a progressive generator which comprises a sequence of transfer blocks. Each block performs an intermediate transfer step by modeling the relationship between the condition and the target poses with attention mechanism. Two types of blocks are introduced, namely pose-attentional transfer block (PATB) and aligned pose-attentional transfer block (APATB). Compared with previous works, our model generates more photorealistic person images that retain better appearance consistency and shape consistency compared with input images. We verify the efficacy of the model on the Market-1501 and DeepFashion datasets, using quantitative and qualitative measures. Furthermore, we show that our method can be used for data augmentation for the person re-identification task, alleviating the issue of data insufficiency. Code and pretrained models are available at: https://github.com/tengteng95/Pose-Transfer.git. [ABSTRACT FROM AUTHOR]

Lung cancer is the most common cause of cancer death worldwide. A timely diagnosis of the pulmonary nodules makes it possible to detect lung cancer in the early stage, and thoracic computed tomography (CT) provides a convenient way to diagnose nodules. However, it is hard even for experienced doctors to distinguish them from the massive CT slices. The currently existing nodule datasets are limited in both scale and category, which is insufficient and greatly restricts its applications. In this paper, we collect the largest and most diverse dataset named PN9 for pulmonary nodule detection by far. Specifically, it contains 8,798 CT scans and 40,439 annotated nodules from 9 common classes. We further propose a slice-aware network (SANet) for pulmonary nodule detection. A slice grouped non-local (SGNL) module is developed to capture long-range dependencies among any positions and any channels of one slice group in the feature map. And we introduce a 3D region proposal network to generate pulmonary nodule candidates with high sensitivity, while this detection stage usually comes with many false positives. Subsequently, a false positive reduction module (FPR) is proposed by using the multi-scale feature maps. To verify the performance of SANet and the significance of PN9, we perform extensive experiments compared with several state-of-the-art 2D CNN-based and 3D CNN-based detection methods. Promising evaluation results on PN9 prove the effectiveness of our proposed SANet. The dataset and source code is available at https://mmcheng.net/SANet/. [ABSTRACT FROM AUTHOR]

Stochastic optimization algorithms have been popular for training deep neural networks. Recently, there emerges a new approach of learning-based optimizer, which has achieved promising performance for training neural networks. However, these black-box learning-based optimizers do not fully take advantage of the experience in human-designed optimizers and heavily rely on learning from meta-training tasks, therefore have limited generalization ability. In this paper, we propose a novel optimizer, dubbed as Variational HyperAdam, which is based on a parametric generalized Adam algorithm, i.e., HyperAdam, in a variational framework. With Variational HyperAdam as optimizer for training neural network, the parameter update vector of the neural network at each training step is considered as random variable, whose approximate posterior distribution given the training data and current network parameter vector is predicted by Variational HyperAdam. The parameter update vector for network training is sampled from this approximate posterior distribution. Specifically, in Variational HyperAdam, we design a learnable generalized Adam algorithm for estimating expectation, paired with a VarBlock for estimating the variance of the approximate posterior distribution of parameter update vector. The Variational HyperAdam is learned in a meta-learning approach with meta-training loss derived by variational inference. Experiments verify that the learned Variational HyperAdam achieved state-of-the-art network training performance for various types of networks on different datasets, such as multilayer perceptron, CNN, LSTM and ResNet. [ABSTRACT FROM AUTHOR]

This paper presents a novel method, Zero-Reference Deep Curve Estimation (Zero-DCE), which formulates light enhancement as a task of image-specific curve estimation with a deep network. Our method trains a lightweight deep network, DCE-Net, to estimate pixel-wise and high-order curves for dynamic range adjustment of a given image. The curve estimation is specially designed, considering pixel value range, monotonicity, and differentiability. Zero-DCE is appealing in its relaxed assumption on reference images, i.e., it does not require any paired or even unpaired data during training. This is achieved through a set of carefully formulated non-reference loss functions, which implicitly measure the enhancement quality and drive the learning of the network. Despite its simplicity, we show that it generalizes well to diverse lighting conditions. Our method is efficient as image enhancement can be achieved by an intuitive and simple nonlinear curve mapping. We further present an accelerated and light version of Zero-DCE, called Zero-DCE++, that takes advantage of a tiny network with just 10K parameters. Zero-DCE++ has a fast inference speed (1000/11 FPS on a single GPU/CPU for an image of size 1200×900×3) while keeping the enhancement performance of Zero-DCE. Extensive experiments on various benchmarks demonstrate the advantages of our method over state-of-the-art methods qualitatively and quantitatively. Furthermore, the potential benefits of our method to face detection in the dark are discussed. The source code is made publicly available at https://li-chongyi.github.io/Proj_Zero-DCE++.html. [ABSTRACT FROM AUTHOR]

Point cloud registration (PCR) is an important and fundamental problem in 3D computer vision, whose goal is to seek an optimal rigid model to register a point cloud pair. Correspondence-based PCR techniques do not require initial guesses and gain more attentions. However, 3D keypoint techniques are much more difficult than their 2D counterparts, which results in extremely high outlier rates. Current robust techniques suffer from very high computational cost. In this paper, we propose a polynomial time ($O(N^2)$ O (N 2) , where $N$ N is the number of correspondences.) outlier removal method. Its basic idea is to reduce the input set into a smaller one with a lower outlier rate based on bound principle. To seek tight lower and upper bounds, we originally define two concepts, i.e., correspondence matrix (CM) and augmented correspondence matrix (ACM). We propose a cost function to minimize the determinant of CM or ACM, where the cost of CM rises to a tight lower bound and the cost of ACM leads to a tight upper bound. Then, we propose a scale-adaptive Cauchy estimator (SA-Cauchy) for further optimization. Extensive experiments on simulated and real PCR datasets demonstrate that the proposed method is robust at outlier rates above 99 percent and 1 $\sim$ ∼ 2 orders faster than its competitors. The source code will be made publicly available in https://ljy-rs.github.io/web/. [ABSTRACT FROM AUTHOR]