CVPR 2019 论文列表

Although Generative Adversarial Networks (GANs) are known for their sharp realism in image generation, they often fail to estimate areas of the data density. This leads to low modal diversity and at times distorted generated samples. These problems essentially arise from poor estimation of the distance metric responsible for training these networks. To address these issues, we introduce an additional regularisation term which performs optimal transport in parallel within a low dimensional representation space. We demonstrate that operating in a low dimension representation of the data distribution benefits from convergence rate gains in estimating the Wasserstein distance, resulting in more stable GAN training. We empirically show that our regulariser achieves a stabilising effect which leads to higher quality of generated samples and increased mode coverage of the given data distribution. Our method achieves significant improvements on the CIFAR-10, Oxford Flowers and CUB Birds datasets over several GAN baselines both qualitatively and quantitatively.

We propose and study a task we name panoptic segmentation (PS). Panoptic segmentation unifies the typically distinct tasks of semantic segmentation (assign a class label to each pixel) and instance segmentation (detect and segment each object instance). The proposed task requires generating a coherent scene segmentation that is rich and complete, an important step toward real-world vision systems. While early work in computer vision addressed related image/scene parsing tasks, these are not currently popular, possibly due to lack of appropriate metrics or associated recognition challenges. To address this, we propose a novel panoptic quality (PQ) metric that captures performance for all classes (stuff and things) in an interpretable and unified manner. Using the proposed metric, we perform a rigorous study of both human and machine performance for PS on three existing datasets, revealing interesting insights about the task. The aim of our work is to revive the interest of the community in a more unified view of image segmentation. For more analysis and up-to-date results, please check the arXiv version of the paper: \smallhttps://arxiv.org/abs/1801.00868 .

The recently introduced panoptic segmentation task has renewed our community's interest in unifying the tasks of instance segmentation (for thing classes) and semantic segmentation (for stuff classes). However, current state-of-the-art methods for this joint task use separate and dissimilar networks for instance and semantic segmentation, without performing any shared computation. In this work, we aim to unify these methods at the architectural level, designing a single network for both tasks. Our approach is to endow Mask R-CNN, a popular instance segmentation method, with a semantic segmentation branch using a shared Feature Pyramid Network (FPN) backbone. Surprisingly, this simple baseline not only remains effective for instance segmentation, but also yields a lightweight, top-performing method for semantic segmentation. In this work, we perform a detailed study of this minimally extended version of Mask R-CNN with FPN, which we refer to as Panoptic FPN, and show it is a robust and accurate baseline for both tasks. Given its effectiveness and conceptual simplicity, we hope our method can serve as a strong baseline and aid future research in panoptic segmentation.

This paper addresses the challenge of 6DoF pose estimation from a single RGB image under severe occlusion or truncation. Many recent works have shown that a two-stage approach, which first detects keypoints and then solves a Perspective-n-Point (PnP) problem for pose estimation, achieves remarkable performance. However, most of these methods only localize a set of sparse keypoints by regressing their image coordinates or heatmaps, which are sensitive to occlusion and truncation. Instead, we introduce a Pixel-wise Voting Network (PVNet) to regress pixel-wise vectors pointing to the keypoints and use these vectors to vote for keypoint locations. This creates a flexible representation for localizing occluded or truncated keypoints. Another important feature of this representation is that it provides uncertainties of keypoint locations that can be further leveraged by the PnP solver. Experiments show that the proposed approach outperforms the state of the art on the LINEMOD, Occlusion LINEMOD and YCB-Video datasets by a large margin, while being efficient for real-time pose estimation. We further create a Truncation LINEMOD dataset to validate the robustness of our approach against truncation. The code is available at https://zju3dv.github.io/pvnet/.

In this paper, we present a novel framework to detect line segments in man-made environments. Specifically, we propose to describe junctions, line segments and relationships between them with a simple graph, which is more structured and informative than end-point representation used in existing line segment detection methods. In order to extract a line segment graph from an image, we further introduce the PPGNet, a convolutional neural network that directly infers a graph from an image. We evaluate our method on published benchmarks including York Urban and Wireframe datasets. The results demonstrate that our method achieves satisfactory performance and generalizes well on all the benchmarks. The source code of our work is available at https://github.com/svip-lab/PPGNet.

In this paper, we proposed a novel haze removal algorithm based on a new feature called the patch map. Conventional patch-based haze removal algorithms (e.g. the Dark Channel prior) usually performs dehazing with a fixed patch size. However, it may produce several problems in recovered results such as oversaturation and color distortion. Therefore, in this paper, we designed an adaptive and automatic patch size selection model called the Patch Map Selection Network (PMS-Net) to select the patch size corresponding to each pixel. This network is designed based on the convolutional neural network (CNN), which can generate the patch map from the image to image. Experimental results on both synthesized and real-world hazy images show that, with the combination of the proposed PMS-Net, the performance in haze removal is much better than that of other state-of-the-art algorithms and we can address the problems caused by the fixed patch size.

The role of pose invariance in image recognition and retrieval is studied. A taxonomic classification of embeddings, according to their level of invariance, is introduced and used to clarify connections between existing embeddings, identify missing approaches, and propose invariant generalizations. This leads to a new family of pose invariant embeddings (PIEs), derived from existing approaches by a combination of two models, which follow from the interpretation of CNNs as estimators of class posterior probabilities: a view-to-object model and an object-to-class model. The new pose-invariant models are shown to have interesting properties, both theoretically and through experiments, where they outperform existing multiview approaches. Most notably, they achieve good performance for both 1) classification and retrieval, and 2) single and multiview inference. These are important properties for the design of real vision systems, where universal embeddings are preferable to task specific ones, and multiple images are usually not available at inference time. Finally, a new multiview dataset of real objects, imaged in the wild against complex backgrounds, is introduced. We believe that this is a much needed complement to the synthetic datasets in wide use and will contribute to the advancement of multiview recognition and retrieval.

Recently, a generative adversarial network (GAN)-based method employing the coarse-to-fine network with the contextual attention module (CAM) has shown outstanding results in image inpainting. However, this method requires numerous computational resources due to its two-stage process for feature encoding. To solve this problem, in this paper, we present a novel network structure, called PEPSI: parallel extended-decoder path for semantic inpainting. PEPSI can reduce the number of convolution operations by adopting a structure consisting of a single shared encoding network and a parallel decoding network with coarse and inpainting paths. The coarse path produces a preliminary inpainting result with which the encoding network is trained to predict features for the CAM. At the same time, the inpainting path creates a higher-quality inpainting result using refined features reconstructed by the CAM. PEPSI not only reduces the number of convolution operation almost by half as compared to the conventional coarse-to-fine networks but also exhibits superior performance to other models in terms of testing time and qualitative scores.

Following the seminal work of Nesterov, accelerated optimization methods have been used to powerfully boost the performance of first-order, gradient-based parameter estimation in scenarios where second-order optimization strategies are either inapplicable or impractical. Accelerated gradient descent converges faster and performs a more robust local search of the parameter space by initially overshooting then oscillating back into minimizers which have a basis of attraction large enough to contain the overshoot. Recent work has demonstrated how a broad class of accelerated schemes can be cast in a variational framework leading to continuum limit ODE's. We extend their formulation to the PDE framework, specifically for the infinite dimensional manifold of continuous curves, to introduce acceleration, and its added robustness, into the broad range of PDE based active contours.

Point cloud based retrieval for place recognition is an emerging problem in vision field. The main challenge is how to find an efficient way to encode the local features into a discriminative global descriptor. In this paper, we propose a Point Contextual Attention Network (PCAN), which can predict the significance of each local point feature based on point context. Our network makes it possible to pay more attention to the task-relevent features when aggregating local features. Experiments on various benchmark datasets show that the proposed network can provide outperformance than current state-of-the-art approaches.

Recent studies have witnessed the successes of using 3D CNNs for video action recognition. However, most 3D models are built upon RGB and optical flow streams, which may not fully exploit pose dynamics, i.e., an important cue of modeling human actions. To fill this gap, we propose a concise Pose-Action 3D Machine (PA3D), which can effectively encode multiple pose modalities within a unified 3D framework, and consequently learn spatio-temporal pose representations for action recognition. More specifically, we introduce a novel temporal pose convolution to aggregate spatial poses over frames. Unlike the classical temporal convolution, our operation can explicitly learn the pose motions that are discriminative to recognize human actions. Extensive experiments on three popular benchmarks (i.e., JHMDB, HMDB, and Charades) show that, PA3D outperforms the recent pose-based approaches. Furthermore, PA3D is highly complementary to the recent 3D CNNs, e.g., I3D. Multi-stream fusion achieves the state-of-the-art performance on all evaluated data sets.

Recently, deep convolutional neural networks (CNNs) have achieved great success in pathological image classification. However, due to the limited number of labeled pathological images, there are still two challenges to be addressed: (1) overfitting: the performance of a CNN model is undermined by the overfitting due to its huge amounts of parameters and the insufficiency of labeled training data. (2) privacy leakage: the model trained using a conventional method may involuntarily reveal the private information of the patients in the training dataset. The smaller the dataset, the worse the privacy leakage. To tackle the above two challenges, we introduce a novel stochastic gradient descent (SGD) scheme, named patient privacy preserving SGD (P3SGD), which performs the model update of the SGD in the patient level via a large-step update built upon each patient's data. Specifically, to protect privacy and regularize the CNN model, we propose to inject the well-designed noise into the updates. Moreover, we equip our P3SGD with an elaborated strategy to adaptively control the scale of the injected noise. To validate the effectiveness of P3SGD, we perform extensive experiments on a real-world clinical dataset and quantitatively demonstrate the superior ability of P3SGD in reducing the risk of overfitting. We also provide a rigorous analysis of the privacy cost under differential privacy. Additionally, we find that the models trained with P3SGD are resistant to the model-inversion attack compared with those trained using non-private SGD.

Cosine-based softmax losses significantly improve the performance of deep face recognition networks. However, these losses always include sensitive hyper-parameters which can make training process unstable, and it is very tricky to set suitable hyper parameters for a specific dataset. This paper addresses this challenge by directly designing the gradients for training in an adaptive manner. We first investigate and unify previous cosine softmax losses from the perspective of gradients. This unified view inspires us to propose a novel gradient called P2SGrad (Probability-to-Similarity Gradient), which leverages a cosine similarity instead of classification probability to control the gradients for updating neural network parameters. P2SGrad is adaptive and hyper-parameter free, which makes training process more efficient and faster. We evaluate our P2SGrad on three face recognition benchmarks, LFW, MegaFace, and IJB-C. The results show that P2SGrad is stable in training, robust to noise, and achieves state-of-the-art performance on all the three benchmarks.

Future prediction is a fundamental principle of intelligence that helps plan actions and avoid possible dangers. As the future is uncertain to a large extent, modeling the uncertainty and multimodality of the future states is of great relevance. Existing approaches are rather limited in this regard and mostly yield a single hypothesis of the future or, at the best, strongly constrained mixture components that suffer from instabilities in training and mode collapse. In this work, we present an approach that involves the prediction of several samples of the future with a winner-takes-all loss and iterative grouping of samples to multiple modes. Moreover, we discuss how to evaluate predicted multimodal distributions, including the common real scenario, where only a single sample from the ground-truth distribution is available for evaluation. We show on synthetic and real data that the proposed approach triggers good estimates of multimodal distributions and avoids mode collapse.

Generalized zero-shot action recognition is a challenging problem, where the task is to recognize new action categories that are unavailable during the training stage, in addition to the seen action categories. Existing approaches suffer from the inherent bias of the learned classifier towards the seen action categories. As a consequence, unseen category samples are incorrectly classified as belonging to one of the seen action categories. In this paper, we set out to tackle this issue by arguing for a separate treatment of seen and unseen action categories in generalized zero-shot action recognition. We introduce an out-of-distribution detector that determines whether the video features belong to a seen or unseen action category. To train our out-of-distribution detector, video features for unseen action categories are synthesized using generative adversarial networks trained on seen action category features. To the best of our knowledge, we are the first to propose an out-of-distribution detector based GZSL framework for action recognition in videos. Experiments are performed on three action recognition datasets: Olympic Sports, HMDB51 and UCF101. For generalized zero-shot action recognition, our proposed approach outperforms the baseline with absolute gains (in classification accuracy) of 7.0%, 3.4%, and 4.9%, respectively, on these datasets.

The weight sharing scheme and spatial pooling operations in Convolutional Neural Networks (CNNs) introduce semantic correlation to neighboring pixels on feature maps and therefore deteriorate their pixel-wise classification performance. In this paper, we implement an Orthogonal Decomposition Unit (ODU) that transforms a convolutional feature map into orthogonal bases targeting at de-correlating neighboring pixels on convolutional features. In theory, complete orthogonal decomposition produces orthogonal bases which can perfectly reconstruct any binary mask (ground-truth). In practice, we further design incomplete orthogonal decomposition focusing on de-correlating local patches which balances the reconstruction performance and computational cost. Fully Convolutional Networks (FCNs) implemented with ODUs, referred to as Orthogonal Decomposition Networks (ODNs), learn de-correlated and complementary convolutional features and fuse such features in a pixel-wise selective manner. Over pixel-wise binary classification tasks for two-dimensional image processing, specifically skeleton detection, edge detection, and saliency detection, and one-dimensional keypoint detection, specifically S-wave arrival time detection for earthquake localization, ODNs consistently improves the state-of-the-arts with significant margins.

Recent advances in matrix completion enable data imputation in full-rank matrices by exploiting low dimensional (nonlinear) latent structure. In this paper, we develop a new model for high rank matrix completion (HRMC), together with batch and online methods to fit the model and out-of-sample extension to complete new data. The method works by (implicitly) mapping the data into a high dimensional polynomial feature space using the kernel trick; importantly, the data occupies a low dimensional subspace in this feature space, even when the original data matrix is of full-rank. The online method can handle streaming or sequential data and adapt to non-stationary latent structure, and enjoys much lower space and time complexity than previous methods for HRMC. For example, the time complexity is reduced from O(n^3) to O(r^3), where n is the number of data points, r is the matrix rank in the feature space, and r<< n. We also provide guidance on sampling rate required for these methods to succeed. Experimental results on synthetic data and motion data validate the performance of the proposed methods.

Data-agnostic quasi-imperceptible perturbations on inputs are known to degrade recognition accuracy of deep convolutional networks severely. This phenomenon is considered to be a potential security issue. Moreover, some results on statistical generalization guarantees indicate that the phenomena can be a key to improve the networks' generalization. However, the characteristics of the shared directions of such harmful perturbations remain unknown. Our primal finding is that convolutional networks are sensitive to the directions of Fourier basis functions. We derived the property by specializing a hypothesis of the cause of the sensitivity, known as the linearity of neural networks, to convolutional networks and empirically validated it. As a byproduct of the analysis, we propose an algorithm to create shift-invariant universal adversarial perturbations available in black-box settings.

This paper addresses the following questions pertaining to the intrinsic dimensionality of any given image representation: (i) estimate its intrinsic dimensionality, (ii) develop a deep neural network based non-linear mapping, dubbed DeepMDS, that transforms the ambient representation to the minimal intrinsic space, and (iii) validate the veracity of the mapping through image matching in the intrinsic space. Experiments on benchmark image datasets (LFW, IJB-C and ImageNet-100) reveal that the intrinsic dimensionality of deep neural network representations is significantly lower than the dimensionality of the ambient features. For instance, SphereFace's 512-dim face representation and ResNet's 512-dim image representation have an intrinsic dimensionality of 16 and 19 respectively. Further, the DeepMDS mapping is able to obtain a representation of significantly lower dimensionality while maintaining discriminative ability to a large extent, 59.75% TAR @ 0.1% FAR in 16-dim vs 71.26% TAR in 512-dim on IJB-C and a Top-1 accuracy of 77.0% at 19-dim vs 83.4% at 512-dim on ImageNet-100.