CVPR 2020 论文列表

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.

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.

We propose a general framework for self-supervised learning of transferable visual representations based on Video-Induced Visual Invariances (VIVI). We consider the implicit hierarchy present in the videos and make use of (i) frame-level invariances (e.g. stability to color and contrast perturbations), (ii) shot/clip-level invariances (e.g. robustness to changes in object orientation and lighting conditions), and (iii) video-level invariances (semantic relationships of scenes across shots/clips), to define a holistic self-supervised loss. Training models using different variants of the proposed framework on videos from the YouTube-8M (YT8M) data set, we obtain state-of-the-art self-supervised transfer learning results on the 19 diverse downstream tasks of the Visual Task Adaptation Benchmark (VTAB), using only 1000 labels per task. We then show how to co-train our models jointly with labeled images, outperforming an ImageNet-pretrained ResNet-50 by 0.8 points with 10x fewer labeled images, as well as the previous best supervised model by 3.7 points using the full ImageNet data set.

The goal of self-supervised learning from images is to construct image representations that are semantically meaningful via pretext tasks that do not require semantic annotations. Many pretext tasks lead to representations that are covariant with image transformations. We argue that, instead, semantic representations ought to be invariant under such transformations. Specifically, we develop Pretext-Invariant Representation Learning (PIRL, pronounced as `pearl') that learns invariant representations based on pretext tasks. We use PIRL with a commonly used pretext task that involves solving jigsaw puzzles. We find that PIRL substantially improves the semantic quality of the learned image representations. Our approach sets a new state-of-the-art in self-supervised learning from images on several popular benchmarks for self-supervised learning. Despite being unsupervised, PIRL outperforms supervised pre-training in learning image representations for object detection. Altogether, our results demonstrate the potential of self-supervised representations with good invariance properties.

Previous methods on estimating detailed human depth often require supervised training with 'ground truth' depth data. This paper presents a self-supervised method that can be trained on YouTube videos without known depth, which makes training data collection simple and improves the generalization of the learned network. The self-supervised learning is achieved by minimizing a photo-consistency loss, which is evaluated between a video frame and its neighboring frames warped according to the estimated depth and the 3D non-rigid motion of the human body. To solve this non-rigid motion, we first estimate a rough SMPL model at each video frame and compute the non-rigid body motion accordingly, which enables self-supervised learning on estimating the shape details. Experiments demonstrate that our method enjoys better generalization, and performs much better on data in the wild.

Generalized Zero-Shot Learning (GZSL) aims at recognizing both seen and unseen classes by constructing correspondence between visual and semantic embedding. However, existing methods have severely suffered from the strong bias problem, where unseen instances in target domain tend to be recognized as seen classes in source domain. To address this issue, we propose an end-to-end Self-supervised Domain-aware Generative Network (SDGN) by integrating self-supervised learning into feature generating model for unbiased GZSL. The proposed SDGN model enjoys several merits. First, we design a cross-domain feature generating module to synthesize samples with high fidelity based on class embeddings, which involves a novel target domain discriminator to preserve the domain consistency. Second, we propose a self-supervised learning module to investigate inter-domain relationships, where a set of anchors are introduced as a bridge between seen and unseen categories. In the shared space, we pull the distribution of target domain away from source domain, and obtain domain-aware features with high discriminative power for both seen and unseen classes. To our best knowledge, this is the first work to introduce self-supervised learning into GZSL as a learning guidance. Extensive experimental results on five standard benchmarks demonstrate that our model performs favorably against state-of-the-art GZSL methods.

In this paper, we look into the problem of 3D adversary attack, and propose to leverage the internal properties of the point clouds and the adversarial examples to design a new self-robust deep neural network (DNN) based 3D recognition systems. As a matter of fact, on one hand, point clouds are highly structured. Hence for each local part of clean point clouds, it is possible to learn what is it ("part of a bottle") and its relative position ("upper part of a bottle") to the global object center. On the other hand, with the visual quality constraint, 3D adversarial samples often only produce small local perturbations, thus they will roughly keep the original global center but may cause incorrect local relative position estimation. Motivated by these two properties, we use relative position (dubbed as "gather-vector") as the adversarial indicator and propose a new robust gather module. Equipped with this module, we further propose a new self-robust 3D point recognition network. Through extensive experiments, we demonstrate that the proposed method can improve the robustness of the target attack under the white-box setting significantly. For I-FGSM based attack, our method reduces the attack success rate from 94.37 % to 75.69 %. For C&W based attack, our method reduces the attack success rate more than 40.00 %. Moreover, our method is complementary to other types of defense methods to achieve better defense results.

In this paper, we solve the sample shortage problem in the human parsing task. We begin with the self-learning strategy, which generates pseudo-labels for unlabeled data to retrain the model. However, directly using noisy pseudo-labels will cause error amplification and accumulation. Considering the topology structure of human body, we propose a trainable graph reasoning method that establishes internal structural connections between graph nodes to correct two typical errors in the pseudo-labels, i.e., the global structural error and the local consistency error. For the global error, we first transform category-wise features into a high-level graph model with coarse-grained structural information, and then decouple the high-level graph to reconstruct the category features. The reconstructed features have a stronger ability to represent the topology structure of the human body. Enlarging the receptive field of features can effectively reducing the local error. We first project feature pixels into a local graph model to capture pixel-wise relations in a hierarchical graph manner, then reverse the relation information back to the pixels. With the global structural and local consistency modules, these errors are rectified and confident pseudo-labels are generated for retraining. Extensive experiments on the LIP and the ATR datasets demonstrate the effectiveness of our global and local rectification modules. Our method outperforms other state-of-the-art methods in supervised human parsing tasks.

In this paper, we address the problem of rain streaks removal in video by developing a self-learned rain streak removal method, which does not require any clean groundtruth images in the training process. The method is inspired by fact that the adjacent frames are highly correlated and can be regarded as different versions of identical scene, and rain streaks are randomly distributed along the temporal dimension. With this in mind, we construct a two-stage Self-Learned Deraining Network (SLDNet) to remove rain streaks based on both temporal correlation and consistency. In the first stage, SLDNet utilizes the temporal correlations and learns to predict the clean version of the current frame based on its adjacent rain video frames. In the second stage, SLDNet enforces the temporal consistency among different frames. It takes both the current rain frame and adjacent rain video frames to recover structural details. The first stage is responsible for reconstructing main structures, and the second stage is responsible for extracting structural details. We build our network architecture with two sub-tasks, i.e. motion estimation, and rain region detection, and optimize them jointly. Our extensive experiments demonstrate the effectiveness of our method, offering better results both quantitatively and qualitatively.

One crucial aspect of partial domain adaptation (PDA) is how to select the relevant source samples in the shared classes for knowledge transfer. Previous PDA methods tackle this problem by re-weighting the source samples based on their high-level information (deep features). However, since the domain shift between source and target domains, only using the deep features for sample selection is defective. We argue that it is more reasonable to additionally exploit the pixel-level information for PDA problem, as the appearance difference between outlier source classes and target classes is significantly large. In this paper, we propose a reinforced transfer network (RTNet), which utilizes both high-level and pixel-level information for PDA problem. Our RTNet is composed of a reinforced data selector (RDS) based on reinforcement learning (RL), which filters out the outlier source samples, and a domain adaptation model which minimizes the domain discrepancy in the shared label space. Specifically, in the RDS, we design a novel reward based on the reconstruct errors of selected source samples on the target generator, which introduces the pixel-level information to guide the learning of RDS. Besides, we develope a state containing high-level information, which used by the RDS for sample selection. The proposed RDS is a general module, which can be easily integrated into existing DA models to make them fit the PDA situation. Extensive experiments indicate that RTNet can achieve state-of-the-art performance for PDA tasks on several benchmark datasets.

Depth data containing a preponderance of discriminative power in location have been proven beneficial for accurate saliency prediction. However, RGB-D saliency detection methods are also negatively influenced by randomly distributed erroneous or missing regions on the depth map or along the object boundaries. This offers the possibility of achieving more effective inference by well designed models. In this paper, we propose a new framework for accurate RGB-D saliency detection taking account of local and global complementarities from two modalities. This is achieved by designing a complimentary interaction model discriminative enough to simultaneously select useful representation from RGB and depth data, and meanwhile to refine the object boundaries. Moreover, we proposed a compensation-aware loss to further process the information not being considered in the complimentary interaction model, leading to improvement of the generalization ability for challenging scenes. Experiments on six public datasets show that our method outperforms18state-of-the-art methods.

Finding a small subset of data whose linear combination spans other data points, also called column subset selection problem (CSSP), is an important open problem in computer science with many applications in computer vision and deep learning. There are some studies that solve CSSP in a polynomial time complexity w.r.t. the size of the original dataset. A simple and efficient selection algorithm with a linear complexity order, referred to as spectrum pursuit (SP), is proposed that pursuits spectral components of the dataset using available sample points. The proposed non-greedy algorithm aims to iteratively find K data samples whose span is close to that of the first K spectral components of entire data. SP has no parameter to be fine tuned and this desirable property makes it problem-independent. The simplicity of SP enables us to extend the underlying linear model to more complex models such as nonlinear manifolds and graph-based models. The nonlinear extension of SP is introduced as kernel-SP (KSP). The superiority of the proposed algorithms is demonstrated in a wide range of applications.

Fuzzy clustering is known to perform well in real-world applications. Inspired by this observation, we incorporate a fuzzy mechanism into discrete convolutional kernels for 3D point clouds as our first major contribution. The proposed fuzzy kernel is defined over a spherical volume that uses discrete bins. Discrete volumetric division can normally make a kernel vulnerable to boundary effects during learning as well as point density during inference. However, the proposed kernel remains robust to boundary conditions and point density due to the fuzzy mechanism. Our second major contribution comes as the proposal of an efficient graph convolutional network, SegGCN for segmenting point clouds. The proposed network exploits ResNet like blocks in the encoder and 1 x 1 convolutions in the decoder. SegGCN capitalizes on the separable convolution operation of the proposed fuzzy kernel for efficiency. We establish the effectiveness of the SegGCN with the proposed kernel on the challenging S3DIS and ScanNet real-world datasets. Our experiments demonstrate that the proposed network can segment over one million points per second with highly competitive performance.

The majority of current object detectors lack context: class predictions are made independently from other detections. We propose to incorporate context in object detection by post-processing the output of an arbitrary detector to rescore the confidences of its detections. Rescoring is done by conditioning on contextual information from the entire set of detections: their confidences, predicted classes, and positions. We show that AP can be improved by simply reassigning the detection confidence values such that true positives that survive longer (i.e., those with the correct class and large IoU) are scored higher than false positives or detections with small IoU. In this setting, we use a bidirectional RNN with attention for contextual rescoring and introduce a training target that uses the IoU with ground truth to maximize AP for the given set of detections. The fact that our approach does not require access to visual features makes it computationally inexpensive and agnostic to the detection architecture. In spite of this simplicity, our model consistently improves AP over strong pre-trained baselines (Cascade R-CNN and Faster R-CNN with several backbones), particularly by reducing the confidence of duplicate detections (a learned form of non-maximum suppression) and removing out-of-context objects by conditioning on the confidences, classes, positions, and sizes of the co-occurrent detections. Code is available at https://github.com/LourencoVazPato/seeing-without-looking/

The fusion of multimodal sensor streams, such as camera, lidar, and radar measurements, plays a critical role in object detection for autonomous vehicles, which base their decision making on these inputs. While existing methods exploit redundant information in good environmental conditions, they fail in adverse weather where the sensory streams can be asymmetrically distorted. These rare "edge-case" scenarios are not represented in available datasets, and existing fusion architectures are not designed to handle them. To address this challenge we present a novel multimodal dataset acquired in over 10,000 km of driving in northern Europe. Although this dataset is the first large multimodal dataset in adverse weather, with 100k labels for lidar, camera, radar, and gated NIR sensors, it does not facilitate training as extreme weather is rare. To this end, we present a deep fusion network for robust fusion without a large corpus of labeled training data covering all asymmetric distortions. Departing from proposal-level fusion, we propose a single-shot model that adaptively fuses features, driven by measurement entropy. We validate the proposed method, trained on clean data, on our extensive validation dataset. Code and data are available here https://github.com/princeton-computational-imaging/SeeingThroughFog.