CVPR 2020 论文列表

In spite of the overwhelming high-tech marvels and applications that rule our digital lives, the use of the handwritten signature is still recognized worldwide in government, personal and legal entities to be the most important behavioral biometric trait. A number of notable research approaches provide advanced results up to a certain point which allow us to assert with confidence that the performance attained by signature verification (SV) systems is comparable to those provided by any other biometric modality. Up to now, the mainstream trend for offline SV is shared between standard -or handcrafted- feature extraction methods and popular machine learning techniques, with typical examples ranging from sparse representation to Deep Learning. Recent progress in graph mining algorithms provide us with the prospect to re-evaluate the opportunity of utilizing graph representations by exploring corresponding graph features for offline SV. In this paper, inspired by the recent use of image visibility graphs for mapping images into networks, we introduce for the first time in offline SV literature their use as a parameter free, agnostic representation for exploring global as well as local information. Global properties of the sparsely located content of the shape of the signature image are encoded with topological information of the whole graph. In addition, local pixel patches are encoded by sequential visibility motifs-subgraphs of size four, to a low six dimensional motif profile vector. A number of pooling functions operate on the motif codes in a spatial pyramid context in order to create the final feature vector. The effectiveness of the proposed method is evaluated with the use of two popular datasets. The local visibility graph features are considered to be highly informative for SV; this is sustained by the corresponding results which are at least comparable with other classic state-of-the-art approaches.

We explore the behavior of a standard convolutional neural net in a continual-learning setting that introduces visual classification tasks sequentially and requires the net to master new tasks while preserving mastery of previously learned tasks. This setting corresponds to that which human learners face as they acquire domain expertise serially, for example, as an individual studies a textbook. Through simulations involving sequences of ten related visual tasks, we find reason for optimism that nets will scale well as they advance from having a single skill to becoming multi-skill domain experts. We observe two key phenomena. First, forward facilitation---the accelerated learning of task n+1 having learned n previous tasks---grows with n. Second, backward interference---the forgetting of the n previous tasks when learning task n+1 ---diminishes with n. Amplifying forward facilitation is the goal of research on metalearning, and attenuating backward interference is the goal of research on catastrophic forgetting. We find that both of these goals are attained simply through broader exposure to a domain.

This work addresses the problem of 3D human pose and shape estimation from a sequence of point clouds. Existing sequential 3D human shape estimation methods mainly focus on the template model fitting from a sequence of depth images or the parametric model regression from a sequence of RGB images. In this paper, we propose a novel sequential 3D human pose and shape estimation framework from a sequence of point clouds. Specifically, the proposed framework can regress 3D coordinates of mesh vertices at different resolutions from the latent features of point clouds. Based on the estimated 3D coordinates and features at the low resolution, we develop a spatial-temporal mesh attention convolution (MAC) to predict the 3D coordinates of mesh vertices at the high resolution. By assigning specific attentional weights to different neighboring points in the spatial and temporal domains, our spatial-temporal MAC can capture structured spatial and temporal features of point clouds. We further generalize our framework to the real data of human bodies with a weakly supervised fine-tuning method. The experimental results on SURREAL, Human3.6M, DFAUST and the real detailed data demonstrate that the proposed approach can accurately recover the 3D body model sequence from a sequence of point clouds.

Particulate matter (PM) is the blend of various solid and liquid particles suspended in atmosphere. These submicron particles are imperceptible for usual hand-held camera photography, but become a great obstacle in microscopic imaging. PM removal from a single microscopic image is a highly ill-posed and one of the challenging image denoising problems. In this work, we thoroughly analyze the physical properties of PM, microscope and their inevitable interaction; and propose an optimization scheme, which removes the PM from a high-resolution microscopic image within a few seconds. Experiments on real world microscopic images show that the proposed method significantly outperforms other competitive image denoising methods. It preserves the comprehensive microscopic foreground details while clearly separating the PM from a single monochromatic or color image.

In this paper, we present a novel cross-consistency based semi-supervised approach for semantic segmentation. Consistency training has proven to be a powerful semi-supervised learning framework for leveraging unlabeled data under the cluster assumption, in which the decision boundary should lie in low-density regions. In this work, we first observe that for semantic segmentation, the low-density regions are more apparent within the hidden representations than within the inputs. We thus propose cross-consistency training, where an invariance of the predictions is enforced over different perturbations applied to the outputs of the encoder. Concretely, a shared encoder and a main decoder are trained in a supervised manner using the available labeled examples. To leverage the unlabeled examples, we enforce a consistency between the main decoder predictions and those of the auxiliary decoders, taking as inputs different perturbed versions of the encoder's output, and consequently, improving the encoder's representations. The proposed method is simple and can easily be extended to use additional training signal, such as image-level labels or pixel-level labels across different domains. We perform an ablation study to tease apart the effectiveness of each component, and conduct extensive experiments to demonstrate that our method achieves state-of-the-art results in several datasets.

Building a large image dataset with high-quality object masks for semantic segmentation is costly and time-consuming. In this paper, we introduce a principled semi-supervised framework that only use a small set of fully supervised images (having semantic segmentation labels and box labels) and a set of images with only object bounding box labels (we call it the weak-set). Our framework trains the primary segmentation model with the aid of an ancillary model that generates initial segmentation labels for the weak-set and a self-correction module that improves the generated labels during training using the increasingly accurate primary model. We introduce two variants of the self-correction module using either linear or convolutional functions. Experiments on the PASCAL VOC 2012 and Cityscape datasets show that our models trained with a small fully supervised set perform similar to, or better than, models trained with a large fully supervised set while requiring 7x less annotation effort.

In the last few years, unpaired image-to-image translation has witnessed Remarkable progress. Although the latest methods are able to generate realistic images, they crucially rely on a large number of labeled images. Recently, some methods have tackled the challenging setting of few-shot image-to-image ranslation, reducing the labeled data requirements for the target domain during inference. In this work, we go one step further and reduce the amount of required labeled data also from the source domain during training. To do so, we propose applying semi-supervised learning via a noise-tolerant pseudo-labeling procedure. We also apply a cycle consistency constraint to further exploit the information from unlabeled images, either from the same dataset or external. Additionally, we propose several structural modifications to facilitate the image translation task under these circumstances. Our semi-supervised method for few-shot image translation, called SEMIT, achieves excellent results on four different datasets using as little as 10% of the source labels, and matches the performance of the main fully-supervised competitor using only 20% labeled data. Our code and models are made public at: https://github.com/yaxingwang/SEMIT.

Skeleton-based human action recognition has attracted great interest thanks to the easy accessibility of the human skeleton data. Recently, there is a trend of using very deep feedforward neural networks to model the 3D coordinates of joints without considering the computational efficiency. In this paper, we propose a simple yet effective semantics-guided neural network (SGN) for skeleton-based action recognition. We explicitly introduce the high level semantics of joints (joint type and frame index) into the network to enhance the feature representation capability. In addition, we exploit the relationship of joints hierarchically through two modules, i.e., a joint-level module for modeling the correlations of joints in the same frame and a framelevel module for modeling the dependencies of frames by taking the joints in the same frame as a whole. A strong baseline is proposed to facilitate the study of this field. With an order of magnitude smaller model size than most previous works, SGN achieves the state-of-the-art performance on the NTU60, NTU120, and SYSU datasets.

In this paper, we focus on semantically multi-modal image synthesis (SMIS) task, namely, generating multi-modal images at the semantic level. Previous work seeks to use multiple class-specific generators, constraining its usage in datasets with a small number of classes. We instead propose a novel Group Decreasing Network (GroupDNet) that leverages group convolutions in the generator and progressively decreases the group numbers of the convolutions in the decoder. Consequently, GroupDNet is armed with much more controllability on translating semantic labels to natural images and has plausible high-quality yields for datasets with many classes. Experiments on several challenging datasets demonstrate the superiority of GroupDNet on performing the SMIS task. We also show that GroupDNet is capable of performing a wide range of interesting synthesis applications. Codes and models are available at: https://github.com/Seanseattle/SMIS.

We present a novel GAN-based model that utilizes the space of deep features learned by a pre-trained classification model. Inspired by classical image pyramid representations, we construct our model as a Semantic Generation Pyramid -- a hierarchical framework which leverages the continuum of semantic information encapsulated in such deep features; this ranges from low level information contained in fine features to high level, semantic information contained in deeper features. More specifically, given a set of features extracted from a reference image, our model generates diverse image samples, each with matching features at each semantic level of the classification model. We demonstrate that our model results in a versatile and flexible framework that can be used in various classic and novel image generation tasks. These include: generating images with a controllable extent of semantic similarity to a reference image, and different manipulation tasks such as semantically-controlled inpainting and compositing; all achieved with the same model, with no further training.

Image manipulation can be considered a special case of image generation where the image to be produced is a modification of an existing image. Image generation and manipulation have been, for the most part, tasks that operate on raw pixels. However, the remarkable progress in learning rich image and object representations has opened the way for tasks such as text-to-image or layout-to-image generation that are mainly driven by semantics. In our work, we address the novel problem of image manipulation from scene graphs, in which a user can edit images by merely applying changes in the nodes or edges of a semantic graph that is generated from the image. Our goal is to encode image information in a given constellation and from there on generate new constellations, such as replacing objects or even changing relationships between objects, while respecting the semantics and style from the original image. We introduce a spatio-semantic scene graph network that does not require direct supervision for constellation changes or image edits. This makes it possible to train the system from existing real-world datasets with no additional annotation effort.

Class-incremental learning of deep networks sequentially increases the number of classes to be classified. During training, the network has only access to data of one task at a time, where each task contains several classes. In this setting, networks suffer from catastrophic forgetting which refers to the drastic drop in performance on previous tasks. The vast majority of methods have studied this scenario for classification networks, where for each new task the classification layer of the network must be augmented with additional weights to make room for the newly added classes. Embedding networks have the advantage that new classes can be naturally included into the network without adding new weights. Therefore, we study incremental learning for embedding networks. In addition, we propose a new method to estimate the drift, called semantic drift, of features and compensate for it without the need of any exemplars. We approximate the drift of previous tasks based on the drift that is experienced by current task data. We perform experiments on fine-grained datasets, CIFAR100 and ImageNet-Subset. We demonstrate that embedding networks suffer significantly less from catastrophic forgetting. We outperform existing methods which do not require exemplars and obtain competitive results compared to methods which store exemplars. Furthermore, we show that our proposed SDC when combined with existing methods to prevent forgetting consistently improves results.

Establishing dense correspondences across semantically similar images is a challenging task. Due to the large intra-class variation and background clutter, two common issues occur in current approaches. First, many pixels in a source image are assigned to one target pixel, i.e., many to one matching. Second, some object pixels are assigned to the background pixels, i.e., background matching. We solve the first issue by global feature matching, which maximizes the total matching correlations between images to obtain a global optimal matching matrix. The row sum and column sum constraints are enforced on the matching matrix to induce a balanced solution, thus suppressing the many to one matching. We solve the second issue by applying a staircase function on the class activation maps to re-weight the importance of pixels into four levels from foreground to background. The whole procedure is combined into a unified optimal transport algorithm by converting the maximization problem to the optimal transport formulation and incorporating the staircase weights into optimal transport algorithm to act as empirical distributions. The proposed algorithm achieves state-of-the-art performance on four benchmark datasets. Notably, a 26% relative improvement is achieved on the large-scale SPair-71k dataset.

In last few years, supervised deep learning has emerged as one powerful tool for image denoising, which trains a denoising network over an external dataset of noisy/clean image pairs. However, the requirement on a high-quality training dataset limits the broad applicability of the denoising networks. Recently, there have been a few works that allow training a denoising network on the set of external noisy images only. Taking one step further, this paper proposes a self-supervised learning method which only uses the input noisy image itself for training. In the proposed method, the network is trained with dropout on the pairs of Bernoulli-sampled instances of the input image, and the result is estimated by averaging the predictions generated from multiple instances of the trained model with dropout. The experiments show that the proposed method not only significantly outperforms existing single-image learning or non-learning methods, but also is competitive to the denoising networks trained on external datasets.

We present a simple self-training method that achieves 88.4% top-1 accuracy on ImageNet, which is 2.0% better than the state-of-the-art model that requires 3.5B weakly labeled Instagram images. On robustness test sets, it improves ImageNet-A top-1 accuracy from 61.0% to 83.7%, reduces ImageNet-C mean corruption error from 45.7 to 28.3, and reduces ImageNet-P mean flip rate from 27.8 to 12.2. To achieve this result, we first train an EfficientNet model on labeled ImageNet images and use it as a teacher to generate pseudo labels on 300M unlabeled images. We then train a larger EfficientNet as a student model on the combination of labeled and pseudo labeled images. We iterate this process by putting back the student as the teacher. During the generation of the pseudo labels, the teacher is not noised so that the pseudo labels are as accurate as possible. However, during the learning of the student, we inject noise such as dropout, stochastic depth and data augmentation via RandAugment to the student so that the student generalizes better than the teacher.

Video anomaly detection is of critical practical importance to a variety of real applications because it allows human attention to be focused on events that are likely to be of interest, in spite of an otherwise overwhelming volume of video. We show that applying self-trained deep ordinal regression to video anomaly detection overcomes two key limitations of existing methods, namely, 1) being highly dependent on manually labeled normal training data; and 2) sub-optimal feature learning. By formulating a surrogate two-class ordinal regression task we devise an end-to-end trainable video anomaly detection approach that enables joint representation learning and anomaly scoring without manually labeled normal/abnormal data. Experiments on eight real-world video scenes show that our proposed method outperforms state-of-the-art methods that require no labeled training data by a substantial margin, and enables easy and accurate localization of the identified anomalies. Furthermore, we demonstrate that our method offers effective human-in-the-loop anomaly detection which can be critical in applications where anomalies are rare and the false-negative cost is high.

Training deep neural networks to estimate the viewpoint of objects requires large labeled training datasets. However, manually labeling viewpoints is notoriously hard, error-prone, and time-consuming. On the other hand, it is relatively easy to mine many unlabeled images of an object category from the internet, e.g., of cars or faces. We seek to answer the research question of whether such unlabeled collections of in-the-wild images can be successfully utilized to train viewpoint estimation networks for general object categories purely via self-supervision. Self-supervision here refers to the fact that the only true supervisory signal that the network has is the input image itself. We propose a novel learning framework which incorporates an analysis-by-synthesis paradigm to reconstruct images in a viewpoint aware manner with a generative network, along with symmetry and adversarial constraints to successfully supervise our viewpoint estimation network. We show that our approach performs competitively to fully-supervised approaches for several object categories like human faces, cars, buses, and trains. Our work opens up further research in self-supervised viewpoint learning and serves as a robust baseline for it. We open-source our code at https://github.com/NVlabs/SSV.

Natural scene understanding is a challenging task, particularly when encountering images of multiple objects that are partially occluded. This obstacle is given rise by varying object ordering and positioning. Existing scene understanding paradigms are able to parse only the visible parts, resulting in incomplete and unstructured scene interpretation. In this paper, we investigate the problem of scene de-occlusion, which aims to recover the underlying occlusion ordering and complete the invisible parts of occluded objects. We make the first attempt to address the problem through a novel and unified framework that recovers hidden scene structures without ordering and amodal annotations as supervisions. This is achieved via Partial Completion Network (PCNet)-mask (M) and -content (C), that learn to recover fractions of object masks and contents, respectively, in a self-supervised manner. Based on PCNet-M and PCNet-C, we devise a novel inference scheme to accomplish scene de-occlusion, via progressive ordering recovery, amodal completion and content completion. Extensive experiments on real-world scenes demonstrate the superior performance of our approach to other alternatives. Remarkably, our approach that is trained in a self-supervised manner achieves comparable results to fully-supervised methods. The proposed scene de-occlusion framework benefits many applications, including high-quality and controllable image manipulation and scene recomposition (see Fig. 1), as well as the conversion of existing modal mask annotations to amodal mask annotations.

Monocular depth estimation has become one of the most studied applications in computer vision, where the most accurate approaches are based on fully supervised learning models. However, the acquisition of accurate and large ground truth data sets to model these fully supervised methods is a major challenge for the further development of the area. Self-supervised methods trained with monocular videos constitute one the most promising approaches to mitigate the challenge mentioned above due to the wide-spread availability of training data. Consequently, they have been intensively studied, where the main ideas explored consist of different types of model architectures, loss functions, and occlusion masks to address non-rigid motion. In this paper, we propose two new ideas to improve self-supervised monocular trained depth estimation: 1) self-attention, and 2) discrete disparity prediction. Compared with the usual localised convolution operation, self-attention can explore a more general contextual information that allows the inference of similar disparity values at non-contiguous regions of the image. Discrete disparity prediction has been shown by fully supervised methods to provide a more robust and sharper depth estimation than the more common continuous disparity prediction, besides enabling the estimation of depth uncertainty. We show that the extension of the state-of-the-art self-supervised monocular trained depth estimator Monodepth2 with these two ideas allows us to design a model that produces the best results in the field in KITTI 2015 and Make3D, closing the gap with respect self-supervised stereo training and fully supervised approaches.