CVPR 2019 论文列表

Learning the embedding space, where semantically similar objects are located close together and dissimilar objects far apart, is a cornerstone of many computer vision applications. Existing approaches usually learn a single metric in the embedding space for all available data points, which may have a very complex non-uniform distribution with different notions of similarity between objects, e.g. appearance, shape, color or semantic meaning. Approaches for learning a single distance metric often struggle to encode all different types of relationships and do not generalize well. In this work, we propose a novel easy-to-implement divide and conquer approach for deep metric learning, which significantly improves the state-of-the-art performance of metric learning. Our approach utilizes the embedding space more efficiently by jointly splitting the embedding space and data into K smaller sub-problems. It divides both, the data and the embedding space into K subsets and learns K separate distance metrics in the non-overlapping subspaces of the embedding space, defined by groups of neurons in the embedding layer of the neural network. The proposed approach increases the convergence speed and improves generalization since the complexity of each sub-problem is reduced compared to the original one. We show that our approach outperforms the state-of-the-art by a large margin in retrieval, clustering and re-identification tasks on CUB200-2011, CARS196, Stanford Online Products, In-shop Clothes and PKU VehicleID datasets. Source code: https://bit.ly/dcesml.

We introduce a novel unsupervised domain adaptation approach for object detection. We aim to alleviate the imperfect translation problem of pixel-level adaptations, and the source-biased discriminativity problem of feature-level adaptations simultaneously. Our approach is composed of two stages, i.e., Domain Diversification (DD) and Multi-domain-invariant Representation Learning (MRL). At the DD stage, we diversify the distribution of the labeled data by generating various distinctive shifted domains from the source domain. At the MRL stage, we apply adversarial learning with a multi-domain discriminator to encourage feature to be indistinguishable among the domains. DD addresses the source-biased discriminativity, while MRL mitigates the imperfect image translation. We construct a structured domain adaptation framework for our learning paradigm and introduce a practical way of DD for implementation. Our method outperforms the state-of-the-art methods by a large margin of 3% 11% in terms of mean average precision (mAP) on various datasets.

In this paper, we propose a new generative model for multi-agent trajectory data, focusing on the case of multi-player sports games. Our model leverages graph neural networks (GNNs) and variational recurrent neural networks (VRNNs) to achieve a permutation equivariant model suitable for sports. On two challenging datasets (basketball and soccer), we show that we are able to produce more accurate forecasts than previous methods. We assess accuracy using various metrics, such as log-likelihood and "best of N" loss, based on N different samples of the future. We also measure the distribution of statistics of interest, such as player location or velocity, and show that the distribution induced by our generative model better matches the empirical distribution of the test set. Finally, we show that our model can perform conditional prediction, which lets us answer counterfactual questions such as "how will the players move differently if A passes the ball to B instead of C?"

This paper proposes the divergence triangle as a framework for joint training of a generator model, energy-based model and inference model. The divergence triangle is a compact and symmetric (anti-symmetric) objective function that seamlessly integrates variational learning, adversarial learning, wake-sleep algorithm, and contrastive divergence in a unified probabilistic formulation. This unification makes the processes of sampling, inference, and energy evaluation readily available without the need for costly Markov chain Monte Carlo methods. Our experiments demonstrate that the divergence triangle is capable of learning (1) an energy-based model with well-formed energy landscape, (2) direct sampling in the form of a generator network, and (3) feed-forward inference that faithfully reconstructs observed as well as synthesized data.

We propose a new geometric regularization principle for reconstructing vector fields based on prior knowledge about their divergence. As one important example of this general idea, we focus on vector fields modelling blood flow pattern that should be divergent in arteries and convergent in veins. We show that this previously ignored regularization constraint can significantly improve the quality of vessel tree reconstruction particularly around bifurcations where non-zero divergence is concentrated. Our divergence prior is critical for resolving (binary) sign ambiguity in flow orientations produced by standard vessel filters, e.g. Frangi. Our vessel tree centerline reconstruction combines divergence constraints with robust curvature regularization. Our unsupervised method can reconstruct complete vessel trees with near-capillary details on synthetic and real 3D volumes.

Shadow detection is an important and challenging task for scene understanding. Despite promising results from recent deep learning based methods. Existing works still struggle with ambiguous cases where the visual appearances of shadow and non-shadow regions are similar (referred to as distraction in our context). In this paper, we propose a Distraction-aware Shadow Detection Network (DSDNet) by explicitly learning and integrating the semantics of visual distraction regions in an end-to-end framework. At the core of our framework is a novel standalone, differentiable Distraction-aware Shadow (DS) module, which allows us to learn distraction-aware, discriminative features for robust shadow detection, by explicitly predicting false positives and false negatives. We conduct extensive experiments on three public shadow detection datasets, SBU, UCF and ISTD, to evaluate our method. Experimental results demonstrate that our model can boost shadow detection performance, by effectively suppressing the detection of false positives and false negatives, achieving state-of-the-art results.

State-of-the-art CNN based recognition models are often computationally prohibitive to deploy on low-end devices. A promising high level approach tackling this limitation is knowledge distillation, which let small student model mimic cumbersome teacher model's output to get improved generalization. However, related methods mainly focus on simple task of classification while do not consider complex tasks like object detection. We show applying the vanilla knowledge distillation to detection model gets minor gain. To address the challenge of distilling knowledge in detection model, we propose a fine-grained feature imitation method exploiting the cross-location discrepancy of feature response. Our intuition is that detectors care more about local near object regions. Thus the discrepancy of feature response on the near object anchor locations reveals important information of how teacher model tends to generalize. We design a novel mechanism to estimate those locations and let student model imitate the teacher on them to get enhanced performance. We first validate the idea on a developed lightweight toy detector which carries simplest notion of current state-of-the-art anchor based detection models on challenging KITTI dataset, our method generates up to 15% boost of mAP for the student model compared to the non-imitated counterpart. We then extensively evaluate the method with Faster R-CNN model under various scenarios with common object detection benchmark of Pascal VOC and COCO, imitation alleviates up to 74% performance drop of student model compared to teacher. Codes released at https://github.com/twangnh/Distilling-Object-Detectors

Person re-identification (Re-ID), for matching pedestrians across non-overlapping camera views, has made great progress in supervised learning with abundant labelled data. However, the scalability problem is the bottleneck for applications in large-scale systems. We consider the scalability problem of Re-ID from three aspects: (1) low labelling cost by reducing label amount, (2) low extension cost by reusing existing knowledge and (3) low testing computation cost by using lightweight models. The requirements render scalable Re-ID a challenging problem. To solve these problems in a unified system, we propose a Multi-teacher Adaptive Similarity Distillation Framework, which requires only a few labelled identities of target domain to transfer knowledge from multiple teacher models to a user-specified lightweight student model without accessing source domain data. We propose the Log-Euclidean Similarity Distillation Loss for Re-ID and further integrate the Adaptive Knowledge Aggregator to select effective teacher models to transfer target-adaptive knowledge. Extensive evaluations show that our method can extend with high scalability and the performance is comparable to the state-of-the-art unsupervised and semi-supervised Re-ID methods.

Due to storage and search efficiency, hashing has become significantly prevalent for nearest neighbor search. Particularly, deep hashing methods have greatly improved the search performance, typically under supervised scenarios. In contrast, unsupervised deep hashing models can hardly achieve satisfactory performance due to the lack of supervisory similarity signals. To address this problem, in this paper, we propose a new deep unsupervised hashing model, called DistilHash, which can learn a distilled data set, where data pairs have confident similarity signals. Specifically, we investigate the relationship between the initial but noisy similarity signals learned from local structures and the semantic similarity labels assigned by the optimal Bayesian classifier. We show that, under a mild assumption, some data pairs, of which labels are consistent with those assigned by the optimal Bayesian classifier, can be potentially distilled. With this understanding, we design a simple but effective method to distill data pairs automatically and further adopt a Bayesian learning framework to learn hashing functions from the distilled data set. Extensive experimental results on three widely used benchmark datasets demonstrate that our method achieves state-of-the-art search performance.

Conventional domain adaptation methods usually resort to deep neural networks or subspace learning to find invariant representations across domains. However, most deep learning methods highly rely on large-size source domains and are computationally expensive to train, while subspace learning methods always have a quadratic time complexity that suffers from the large domain size. This paper provides a simple and efficient solution, which could be regarded as a well-performing baseline for domain adaptation tasks. Our method is built upon the nearest centroid classifier, seeking a subspace where the centroids in the target domain are moderately shifted from those in the source domain. Specifically, we design a unified objective without accessing the source domain data and adopt an alternating minimization scheme to iteratively discover the pseudo target labels, invariant subspace, and target centroids. Besides its privacy-preserving property (distant supervision), the algorithm is provably convergent and has a promising linear time complexity. In addition, the proposed method can be readily extended to multi-source setting and domain generalization, and it remarkably enhances popular deep adaptation methods by borrowing the learned transferable features. Extensive experiments on several benchmarks including object, digit, and face recognition datasets validate that our methods yield state-of-the-art results in various domain adaptation tasks.

We consider the problem of weakly supervised object detection, where the training samples are annotated using only image-level labels that indicate the presence or absence of an object category. In order to model the uncertainty in the location of the objects, we employ a dissimilarity coefficient based probabilistic learning objective. The learning objective minimizes the difference between an annotation agnostic prediction distribution and an annotation aware conditional distribution. The main computational challenge is the complex nature of the conditional distribution, which consists of terms over hundreds or thousands of variables. The complexity of the conditional distribution rules out the possibility of explicitly modeling it. Instead, we exploit the fact that deep learning frameworks rely on stochastic optimization. This allows us to use a state of the art discrete generative model that can provide annotation consistent samples from the conditional distribution. Extensive experiments on PASCAL VOC 2007 and 2012 data sets demonstrate the efficacy of our proposed approach.

Variations in visual factors such as viewpoint, pose, illumination and background, are usually viewed as important challenges in person re-identification (re-ID). In spite of acknowledging these factors to be influential, quantitative studies on how they affect a re-ID system are still lacking. To derive insights in this scientific campaign, this paper makes an early attempt in studying a particular factor, viewpoint. We narrow the viewpoint problem down to the pedestrian rotation angle to obtain focused conclusions. In this regard, this paper makes two contributions to the community. First, we introduce a large-scale synthetic data engine, PersonX. Composed of hand-crafted 3D person models, the salient characteristic of this engine is "controllable". That is, we are able to synthesize pedestrians by setting the visual variables to arbitrary values. Second, on the 3D data engine, we quantitatively analyze the influence of pedestrian rotation angle on re-ID accuracy. Comprehensively, the person rotation angles are precisely customized from 0 to 360, allowing us to investigate its effect on the training, query, and gallery sets. Extensive experiment helps us have a deeper understanding of the fundamental problems in person re-ID. Our research also provides useful insights for dataset building and future practical usage, e.g., a person of a side view makes a better query.

VAE requires the standard Gaussian distribution as a prior in the latent space. Since all codes tend to follow the same prior, it often suffers the so-called "posterior collapse". To avoid this, this paper introduces the class specific distribution for the latent code. But different from CVAE, we present a method for disentangling the latent space into the label relevant and irrelevant dimensions, zs and zu, for a single input. We apply two separated encoders to map the input into zs and zu respectively, and then give the concatenated code to the decoder to reconstruct the input. The label irrelevant code zu represent the common characteristics of all inputs, hence they are constrained by the standard Gaussian, and their encoder is trained in amortized variational inference way, like VAE. While zs is assumed to follow the Gaussian mixture distribution in which each component corresponds to a particular class. The parameters for the Gaussian components in zs encoder are optimized by the label supervision in a global stochastic way. In theory, we show that our method is actually equivalent to adding a KL divergence term on the joint distribution of zs and the class label c, and it can directly increase the mutual information between zs and the label c. Our model can also be extended to GAN by adding a discriminator in the pixel domain so that it produces high quality and diverse images.

Hand image synthesis and pose estimation from RGB images are both highly challenging tasks due to the large discrepancy between factors of variation ranging from image background content to camera viewpoint. To better analyze these factors of variation, we propose the use of disentangled representations and a disentangled variational autoencoder (dVAE) that allows for specific sampling and inference of these factors. The derived objective from the variational lower bound as well as the proposed training strategy are highly flexible, allowing us to handle crossmodal encoders and decoders as well as semi-supervised learning scenarios. Experiments show that our dVAE can synthesize highly realistic images of the hand specifiable by both pose and image background content and also estimate 3D hand poses from RGB images with accuracy competitive with state-of-the-art on two public benchmarks.

Obtaining deep networks that are robust against adversarial examples and generalize well is an open problem. A recent hypothesis even states that both robust and accurate models are impossible, i.e., adversarial robustness and generalization are conflicting goals. In an effort to clarify the relationship between robustness and generalization, we assume an underlying, low-dimensional data manifold and show that: 1. regular adversarial examples leave the manifold; 2. adversarial examples constrained to the manifold, i.e., on-manifold adversarial examples, exist; 3. on-manifold adversarial examples are generalization errors, and on-manifold adversarial training boosts generalization; 4. regular robustness and generalization are not necessarily contradicting goals. These assumptions imply that both robust and accurate models are possible. However, different models (architectures, training strategies etc.) can exhibit different robustness and generalization characteristics. To confirm our claims, we present extensive experiments on synthetic data (with known manifold) as well as on EMNIST, Fashion-MNIST and CelebA.

In this paper, we present a novel strategy to design disentangled 3D face shape representation. Specifically, a given 3D face shape is decomposed into identity part and expression part, which are both encoded and decoded in a nonlinear way. To solve this problem, we propose an attribute decomposition framework for 3D face mesh. To better represent face shapes which are usually nonlinear deformed between each other, the face shapes are represented by a vertex based deformation representation rather than Euclidean coordinates. The experimental results demonstrate that our method has better performance than existing methods on decomposing the identity and expression parts. Moreover, more natural expression transfer results can be achieved with our method than existing methods.

Our goal in this paper is to discover near duplicate patterns in large collections of artworks. This is harder than standard instance mining due to differences in the artistic media (oil, pastel, drawing, etc), and imperfections inherent in the copying process. Our key technical insight is to adapt a standard deep feature to this task by fine-tuning it on the specific art collection using self-supervised learning. More specifically, spatial consistency between neighbouring feature matches is used as supervisory fine-tuning signal. The adapted feature leads to more accurate style invariant matching, and can be used with a standard discovery approach, based on geometric verification, to identify duplicate patterns in the dataset. The approach is evaluated on several different datasets and shows surprisingly good qualitative discovery results. For quantitative evaluation of the method, we annotated 273 near duplicate details in a dataset of 1587 artworks attributed to Jan Brueghel and his workshop. Beyond artworks, we also demonstrate improvement on localization on the Oxford5K photo dataset as well as on historical photograph localization on the Large Time Lags Location (LTLL) dataset.

Interpretability and fairness are critical in computer vision and machine learning applications, in particular when dealing with human outcomes, e.g. inviting or not inviting for a job interview based on application materials that may include photographs. One promising direction to achieve fairness is by learning data representations that remove the semantics of protected characteristics, and are therefore able to mitigate unfair outcomes. All available models however learn latent embeddings which comes at the cost of being uninterpretable. We propose to cast this problem as data-to-data translation, i.e. learning a mapping from an input domain to a fair target domain, where a fairness definition is being enforced. Here the data domain can be images, or any tabular data representation. This task would be straightforward if we had fair target data available, but this is not the case. To overcome this, we learn a highly unconstrained mapping by exploiting statistics of residuals -- the difference between input data and its translated version -- and the protected characteristics. When applied to the CelebA dataset of face images with gender attribute as the protected characteristic, our model enforces equality of opportunity by adjusting the eyes and lips regions. Intriguingly, on the same dataset we arrive at similar conclusions when using semantic attribute representations of images for translation. On face images of the recent DiF dataset, with the same gender attribute, our method adjusts nose regions. In the Adult income dataset, also with protected gender attribute, our model achieves equality of opportunity by, among others, obfuscating the wife and husband relationship. Analyzing those systematic changes will allow us to scrutinize the interplay of fairness criterion, chosen protected characteristics, and prediction performance.

Visual object recognition under situations in which the direct line-of-sight is blocked, such as when it is occluded around the corner, is of practical importance in a wide range of applications. With coherent illumination, the light scattered from diffusive walls forms speckle patterns that contain information of the hidden object. It is possible to realize non-line-of-sight (NLOS) recognition with these speckle patterns. We introduce a novel approach based on speckle pattern recognition with deep neural network, which is simpler and more robust than other NLOS recognition methods. Simulations and experiments are performed to verify the feasibility and performance of this approach.