CVPR 2017 Papers

Rain streaks removal is an important issue of the outdoor vision system and has been recently investigated extensively. In this paper, we propose a novel tensor based video rain streaks removal approach by fully considering the discriminatively intrinsic characteristics of rain streaks and clean videos, which needs neither rain detection nor time-consuming dictionary learning stage. In specific, on the one hand, rain streaks are sparse and smooth along the raindrops' direction, and on the other hand, the clean videos possess smoothness along the rain-perpendicular direction and global and local correlation along time direction. We use the l_1 norm to enhance the sparsity of the underlying rain, two unidirectional Total Variation (TV) regularizers to guarantee the different discriminative smoothness, and a tensor nuclear norm and a time directional difference operator to characterize the exclusive correlation of the clean video along time. Alternation direction method of multipliers (ADMM) is employed to solve the proposed concise tensor based convex model. Experiments implemented on synthetic and real data substantiate the effectiveness and efficiency of the proposed method. Under comprehensive quantitative performance measures, our approach outperforms other state-of-the-art methods.

`Speckle' refers to the granular patterns that occur in ultrasound images due to wave interference. Speckle removal can greatly improve the visibility of the underlying structures in an ultrasound image and enhance subsequent post processing. We present a novel framework for speckle removal based on low-rank non-local filtering. Our approach works by first computing a guidance image that assists in the selection of candidate patches for non-local filtering in the face of significant speckles. The candidate patches are further refined using a low-rank minimization estimated using a truncated weighted nuclear norm (TWNN) and structured sparsity. We show that the proposed filtering framework produces results that outperform state-of-the-art methods both qualitatively and quantitatively. This framework also provides better segmentation results when used for pre-processing ultrasound images.

This paper tackles the photometric stereo problem in the presence of inaccurate lighting, obtained either by calibration or by an uncalibrated photometric stereo method. Based on a precise modeling of noise and outliers, a robust variational approach is introduced. It explicitly accounts for self-shadows, and enforces robustness to cast-shadows and specularities by resorting to redescending M-estimators. The resulting non-convex model is solved by means of a computationally efficient alternating reweighted least-squares algorithm. Since it implicitly enforces integrability, the new variational approach can refine both the intensities and the directions of the lighting.

This paper presents a new method for 3D action recognition with skeleton sequences (i.e., 3D trajectories of human skeleton joints). The proposed method first transforms each skeleton sequence into three clips each consisting of several frames for spatial temporal feature learning using deep neural networks. Each clip is generated from one channel of the cylindrical coordinates of the skeleton sequence. Each frame of the generated clips represents the temporal information of the entire skeleton sequence, and incorporates one particular spatial relationship between the joints. The entire clips include multiple frames with different spatial relationships, which provide useful spatial structural information of the human skeleton. We propose to use deep convolutional neural networks to learn long-term temporal information of the skeleton sequence from the frames of the generated clips, and then use a Multi-Task Learning Network (MTLN) to jointly process all frames of the clips in parallel to incorporate spatial structural information for action recognition. Experimental results clearly show the effectiveness of the proposed new representation and feature learning method for 3D action recognition.

Motivated by the limitations of existing multi-view stereo benchmarks, we present a novel dataset for this task. Towards this goal, we recorded a variety of indoor and outdoor scenes using a high-precision laser scanner and captured both high-resolution DSLR imagery as well as synchronized low-resolution stereo videos with varying fields-of-view. To align the images with the laser scans, we propose a robust technique which minimizes photometric errors conditioned on the geometry. In contrast to previous datasets, our benchmark provides novel challenges and covers a diverse set of viewpoints and scene types, ranging from natural scenes to man-made indoor and outdoor environments. Furthermore, we provide data at significantly higher temporal and spatial resolution. Our benchmark is the first to cover the important use case of hand-held mobile devices while also providing high-resolution DSLR camera images. We make our datasets and an online evaluation server available at http://www.eth3d.net.

A minimal solution using two affine correspondences is presented to estimate the common focal length and the fundamental matrix between two semi-calibrated cameras - known intrinsic parameters except a common focal length. To the best of our knowledge, this problem is unsolved. The proposed approach extends point correspondence-based techniques with linear constraints derived from local affine transformations. The obtained multivariate polynomial system is efficiently solved by the hidden-variable technique. Observing the geometry of local affinities, we introduce novel conditions eliminating invalid roots. To select the best one out of the remaining candidates, a root selection technique is proposed outperforming the recent ones especially in case of high-level noise. The proposed 2-point algorithm is validated on both synthetic data and 104 publicly available real image pairs. A Matlab implementation of the proposed solution is included in the paper.

We propose a dual decomposition and linear program relaxation of the NP-hard minimum cost multicut problem. Unlike other polyhedral relaxations of the multicut polytope, it is amenable to efficient optimization by message passing. Like other polyhedral relaxations, it can be tightened efficiently by cutting planes. We define an algorithm that alternates between message passing and efficient separation of cycle- and odd-wheel inequalities. This algorithm is more efficient than state-of-the-art algorithms based on linear programming, including algorithms written in the framework of leading commercial software, as we show in experiments with large instances of the problem from applications in computer vision, biomedical image analysis and data mining.

Composite function minimization captures a wide spectrum of applications in both computer vision and machine learning. It includes bound constrained optimization and cardinality regularized optimization as special cases. This paper proposes and analyzes a new Matrix Splitting Method (MSM) for minimizing composite functions. It can be viewed as a generalization of the classical Gauss-Seidel method and the Successive Over-Relaxation method for solving linear systems in the literature. Incorporating a new Gaussian elimination procedure, the matrix splitting method achieves state-of-the-art performance. For convex problems, we establish the global convergence, convergence rate, and iteration complexity of MSM, while for non-convex problems, we prove its global convergence. Finally, we validate the performance of our matrix splitting method on two particular applications: nonnegative matrix factorization and cardinality regularized sparse coding. Extensive experiments show that our method outperforms existing composite function minimization techniques in term of both efficiency and efficacy.

We present the first gesture recognition system implemented end-to-end on event-based hardware, using a TrueNorth neurosynaptic processor to recognize hand gestures in real-time at low power from events streamed live by a Dynamic Vision Sensor (DVS). The biologically inspired DVS transmits data only when a pixel detects a change, unlike traditional frame-based cameras which sample every pixel at a fixed frame rate. This sparse, asynchronous data representation lets event-based cameras operate at much lower power than frame-based cameras. However, much of the energy efficiency is lost if, as in previous work, the event stream is interpreted by conventional synchronous processors. Here, for the first time, we process a live DVS event stream using TrueNorth, a natively event-based processor with 1 million spiking neurons. Configured here as a convolutional neural network (CNN), the TrueNorth chip identifies the onset of a gesture with a latency of 105 ms while consuming less than 200 mW. The CNN achieves 96.5% out-of-sample accuracy on a newly collected DVS dataset (DvsGesture) comprising 11 hand gesture categories from 29 subjects under 3 illumination conditions.

This paper proposes a new extrinsic calibration of kaleidoscopic imaging system by estimating normals and distances of the mirrors. The problem to be solved in this paper is a simultaneous estimation of all mirror parameters consistent throughout multiple reflections. Unlike conventional methods utilizing a pair of direct and mirrored images of a reference 3D object to estimate the parameters on a per-mirror basis, our method renders the simultaneous estimation problem into solving a linear set of equations. The key contribution of this paper is to introduce a linear estimation of multiple mirror parameters from kaleidoscopic 2D projections of a single 3D point of unknown geometry. Evaluations with synthesized and real images demonstrate the performance of the proposed algorithm in comparison with conventional methods.

Deep Auto-Encoder (DAE) has shown its promising power in high-level representation learning. From the perspective of manifold learning, we propose a graph regularized deep neural network (GR-DNN) to endue traditional DAEs with the ability of retaining local geometric structure. A deep-structured regularizer is formulated upon multi-layer perceptions to capture this structure. The robust and discriminative embedding space is learned to simultaneously preserve the high-level semantics and the geometric structure within local manifold tangent space. Theoretical analysis presents the close relationship between the proposed graph regularizer and the graph Laplacian regularizer in terms of the optimization objective. We also alleviate the growth of the network complexity by introducing the anchor-based bipartite graph, which guarantees the good scalability for large scale data. The experiments on four datasets show the comparable results of the proposed GR-DNN with the state-of-the-art methods.

We introduce a novel technique for knowledge transfer, where knowledge from a pretrained deep neural network (DNN) is distilled and transferred to another DNN. As the DNN performs a mapping from the input space to the output space through many layers sequentially, we define the distilled knowledge to be transferred in terms of flow between layers, which is calculated by computing the inner product between features from two layers. When we compare the student DNN and the original network with the same size as the student DNN but trained without a teacher network, the proposed method of transferring the distilled knowledge as the flow between two layers exhibits three important phenomena : (1) the student DNN that learns the distilled knowledge is optimized much faster than the original model; (2) the student DNN outperforms the original DNN; and (3) the student DNN can learn the distilled knowledge from a teacher DNN that is trained at a different task, and the student DNN outperforms the original DNN that is trained from scratch.

We consider the problem of depth-based robust 3D facial pose tracking under unconstrained scenarios with heavy occlusions and arbitrary facial expression variations. Unlike the previous depth-based discriminative or data-driven methods that require sophisticated training or manual intervention, we propose a generative framework that unifies pose tracking and face model adaptation on-the-fly. Particularly, we propose a statistical 3D face model that owns the flexibility to generate and predict the distribution and uncertainty underlying the face model. Moreover, unlike prior arts employing the ICP-based facial pose estimation, we propose a ray visibility constraint that regularizes the pose based on the face model's visibility against the input point cloud, which augments the robustness against the occlusions. The experimental results on Biwi and ICT-3DHP datasets reveal that the proposed framework is effective and outperforms the state-of-the-art depth-based methods.

This paper introduces a general setting for the construction of data fidelity metrics between oriented or non-oriented geometric shapes like curves, curve sets or surfaces. These metrics are based on the representation of shapes as distributions of their local tangent or normal vectors and the definition of reproducing kernels on these spaces. The construction, that combines in one common setting and extends the previous frameworks of currents and varifolds, provides a very large class of kernel metrics which can be easily computed without requiring any kind of parametrization of shapes and which are smooth enough to give robustness to certain imperfections that could result e.g. from bad segmentation. We then give a sense, with synthetic examples, of the versatility and potentialities of such metrics when used in various problems like shape comparison, clustering and diffeomorphic registration.

We propose a general dual ascent (message passing) framework for Lagrangean (dual) decomposition of combinatorial problems. Although methods of this type have shown their efficiency for a number of problems, so far there was no general algorithm applicable to multiple problem types. In this work, we propose such a general algorithm. It depends on several parameters, which can be used to optimize its performance in each particular setting. We demonstrate efficiency of our method on the graph matching and the multicut problems, where it outperforms state-of-the-art solvers including those based on the subgradient optimization and off-the-shelf linear programming solvers.

Social relations are the foundation of human daily life. Developing techniques to analyze such relations from visual data bears great potential to build machines that better understand us and are capable of interacting with us at a social level. Previous investigations have remained partial due to the overwhelming diversity and complexity of the topic and consequently have only focused on a handful of social relations. In this paper, we argue that the domain-based theory from social psychology is a great starting point to systematically approach this problem. The theory provides coverage of all aspects of social relations and equally is concrete and predictive about the visual attributes and behaviors defining the relations included in each domain. We provide the first dataset built on this holistic conceptualization of social life that is composed of a hierarchical label space of social domains and social relations. We also contribute the first models to recognize such domains and relations and find superior performance for attribute based features. Beyond the encouraging performance of the attribute based approach, we also find interpretable features that are in accordance with the predictions from social psychology literature. Beyond our findings, we believe that our contributions more tightly interleave visual recognition and social psychology theory that has the potential to complement the theoretical work in the area with empirical and data-driven models of social life.

Regression based facial landmark detection methods usually learns a series of regression functions to update the landmark positions from an initial estimation. Most of existing approaches focus on learning effective mapping functions with robust image features to improve performance. The approach to dealing with the initialization issue, however, receives relatively fewer attentions. In this paper, we present a deep regression architecture with two-stage re-initialization to explicitly deal with the initialization problem. At the global stage, given an image with a rough face detection result, the full face region is firstly re-initialized by a supervised spatial transformer network to a canonical shape state and then trained to regress a coarse landmark estimation. At the local stage, different face parts are further separately re-initialized to their own canonical shape states, followed by another regression subnetwork to get the final estimation. Our proposed deep architecture is trained from end to end and obtains promising results using different kinds of unstable initialization. It also achieves superior performances over many competing algorithms.

A novel dataset for benchmarking image-based localization is presented. With increasing research interests in visual place recognition and localization, several datasets have been published in the past few years. One of the evident limitations of existing datasets is that precise ground truth camera poses of query images are not available in a meaningful 3D metric system. This is in part due to the underlying 3D models of these datasets are reconstructed from Structure from Motion methods. So far little attention has been paid to metric evaluations of localization accuracy. In this paper we address the problem of whether state-of-the-art visual localization techniques can be applied to tasks with demanding accuracy requirements. We acquired training data for a large indoor environment with cameras and a LiDAR scanner. In addition, we collected over 2000 query images with cell phone cameras. Using LiDAR point clouds as a reference, we employed a semi-automatic approach to estimate the 6 degrees of freedom camera poses precisely in the world coordinate system. The proposed dataset enables us to quantitatively assess the performance of various algorithms using a fair and intuitive metric.

While deep convolutional neural networks frequently approach or exceed human-level performance in benchmark tasks involving static images, extending this success to moving images is not straightforward. Video understanding is of interest for many applications, including content recommendation, prediction, summarization, event/object detection, and understanding human visual perception. However, many domains lack sufficient data to explore and perfect video models. In order to address the need for a simple, quantitative benchmark for developing and understanding video, we present MovieFIB, a fill-in-the-blank question-answering dataset with over 300,000 examples, based on descriptive video annotations for the visually impaired. In addition to presenting statistics and a description of the dataset, we perform a detailed analysis of 5 different models' predictions, and compare these with human performance. We investigate the relative importance of language, static (2D) visual features, and moving (3D) visual features; the effects of increasing dataset size, the number of frames sampled; and of vocabulary size. We illustrate that: this task is not solvable by a language model alone; our model combining 2D and 3D visual information indeed provides the best result; all models perform significantly worse than human-level. We provide human evaluation for responses given by different models and find that accuracy on the MovieFIB evaluation corresponds well with human judgment. We suggest avenues for improving video models, and hope that the MovieFIB challenge can be useful for measuring and encouraging progress in this very interesting field.