CVPR 2020 论文列表

Color and structure are the two pillars that construct an image. Usually, the structure is well expressed through a rich spectrum of colors, allowing objects in an image to be recognized by neural networks. However, under extreme limitations of color space, the structure tends to vanish, and thus a neural network might fail to understand the image. Interested in exploring this interplay between color and structure, we study the scientific problem of identifying and preserving the most informative image structures while constraining the color space to just a few bits, such that the resulting image can be recognized with possibly high accuracy. To this end, we propose a color quantization network, ColorCNN, which learns to structure the images from the classification loss in an end-to-end manner. Given a color space size, ColorCNN quantizes colors in the original image by generating a color index map and an RGB color palette. Then, this color-quantized image is fed to a pre-trained task network to evaluate its performance. In our experiment, with only a 1-bit color space (i.e., two colors), the proposed network achieves 82.1% top-1 accuracy on the CIFAR10 dataset, outperforming traditional color quantization methods by a large margin. For applications, when encoded with PNG, the proposed color quantization shows superiority over other image compression methods in the extremely low bit-rate regime. The code is available at https://github.com/hou-yz/color_distillation.

Simulation is a crucial component of any robotic system. In order to simulate correctly, we need to write complex rules of the environment: how dynamic agents behave, and how the actions of each of the agents affect the behavior of others. In this paper, we aim to learn a simulator by simply watching an agent interact with an environment. We focus on graphics games as a proxy of the real environment. We introduce GameGAN, a generative model that learns to visually imitate a desired game by ingesting screenplay and keyboard actions during training. Given a key pressed by the agent, GameGAN "renders" the next screen using a carefully designed generative adversarial network. Our approach offers key advantages over existing work: we design a memory module that builds an internal map of the environment, allowing for the agent to return to previously visited locations with high visual consistency. In addition, GameGAN is able to disentangle static and dynamic components within an image making the behavior of the model more interpretable, and relevant for downstream tasks that require explicit reasoning over dynamic elements. This enables many interesting applications such as swapping different components of the game to build new games that do not exist. We will release the code and trained model, enabling human players to play generated games and their variations with our GameGAN.

We present a fully automatic method to generate detailed and accurate artistic shadows from pairs of line drawing sketches and lighting directions. We also contribute a new dataset of one thousand examples of pairs of line drawings and shadows that are tagged with lighting directions. Remarkably, the generated shadows quickly communicate the underlying 3D structure of the sketched scene. Consequently, the shadows generated by our approach can be used directly or as an excellent starting point for artists. We demonstrate that the deep learning network we propose takes a hand-drawn sketch, builds a 3D model in latent space, and renders the resulting shadows. The generated shadows respect the hand-drawn lines and underlying 3D space and contain sophisticated and accurate details, such as self-shadowing effects. Moreover, the generated shadows contain artistic effects, such as rim lighting or halos appearing from backlighting, that would be achievable with traditional 3D rendering methods.

Few-shot learning has attracted intensive research attention in recent years. Many methods have been proposed to generalize a model learned from provided base classes to novel classes, but no previous work studies how to select base classes, or even whether different base classes will result in different generalization performance of the learned model. In this paper, we utilize a simple yet effective measure, the Similarity Ratio, as an indicator for the generalization performance of a few-shot model. We then formulate the base class selection problem as a submodular optimization problem over Similarity Ratio. We further provide theoretical analysis on the optimization lower bound of different optimization methods, which could be used to identify the most appropriate algorithm for different experimental settings. The extensive experiments on ImageNet, Caltech256 and CUB-200-2011 demonstrate that our proposed method is effective in selecting a better base dataset.

Real-world visual recognition requires handling the extreme sample imbalance in large-scale long-tailed data. We propose a "divide&conquer" strategy for the challenging LVIS task: divide the whole data into balanced parts and then apply incremental learning to conquer each one. This derives a novel learning paradigm: class-incremental few-shot learning, which is especially effective for the challenge evolving over time: 1) the class imbalance among the old class knowledge review and 2) the few-shot data in new-class learning. We call our approach Learning to Segment the Tail (LST). In particular, we design an instance-level balanced replay scheme, which is a memory-efficient approximation to balance the instance-level samples from the old-class images. We also propose to use a meta-module for new-class learning, where the module parameters are shared across incremental phases, gaining the learning-to-learn knowledge incrementally, from the data-rich head to the data-poor tail. We empirically show that: at the expense of a little sacrifice of head-class forgetting, we can gain a significant 8.3% AP improvement for the tail classes with less than 10 instances, achieving an overall 2.0% AP boost for the whole 1,230 classes.

In contrast to the literature where local patterns in 3D point clouds are captured by customized convolutional operators, in this paper we study the problem of how to effectively and efficiently project such point clouds into a 2D image space so that traditional 2D convolutional neural networks (CNNs) such as U-Net can be applied for segmentation. To this end, we are motivated by graph drawing and reformulate it as an integer programming problem to learn the topology-preserving graph-to-grid mapping for each individual point cloud. To accelerate the computation in practice, we further propose a novel hierarchical approximate algorithm. With the help of the Delaunay triangulation for graph construction from point clouds and a multi-scale U-Net for segmentation, we manage to demonstrate the state-of-the-art performance on ShapeNet and PartNet, respectively, with significant improvement over the literature. Code is available at https://github.com/Zhang-VISLab.

We present a learning-based approach for removing unwanted obstructions, such as window reflections, fence occlusions or raindrops, from a short sequence of images captured by a moving camera. Our method leverages the motion differences between the background and the obstructing elements to recover both layers. Specifically, we alternate between estimating dense optical flow fields of the two layers and reconstructing each layer from the flow-warped images via a deep convolutional neural network. The learning-based layer reconstruction allows us to accommodate potential errors in the flow estimation and brittle assumptions such as brightness consistency. We show that training on synthetically generated data transfers well to real images. Our results on numerous challenging scenarios of reflection and fence removal demonstrate the effectiveness of the proposed method.

Low-light images typically suffer from two problems. First, they have low visibility (i.e., small pixel values). Second, noise becomes significant and disrupts the image content, due to low signal-to-noise ratio. Most existing lowlight image enhancement methods, however, learn from noise-negligible datasets. They rely on users having good photographic skills in taking images with low noise. Unfortunately, this is not the case for majority of the low-light images. While concurrently enhancing a low-light image and removing its noise is ill-posed, we observe that noise exhibits different levels of contrast in different frequency layers, and it is much easier to detect noise in the lowfrequency layer than in the high one. Inspired by this observation, we propose a frequency-based decompositionand- enhancement model for low-light image enhancement. Based on this model, we present a novel network that first learns to recover image objects in the low-frequency layer and then enhances high-frequency details based on the recovered image objects. In addition, we have prepared a new low-light image dataset with real noise to facilitate learning. Finally, we have conducted extensive experiments to show that the proposed method outperforms state-of-the-art approaches in enhancing practical noisy low-light images.

Many tasks in computer vision and machine learning are modeled as optimization problems with constraints in the form of Symmetric Positive Definite (SPD) matrices. Solving such optimization problems is challenging due to the non-linearity of the SPD manifold, making optimization with SPD constraints heavily relying on expert knowledge and human involvement. In this paper, we propose a meta-learning method to automatically learn an iterative optimizer on SPD manifolds. Specifically, we introduce a novel recurrent model that takes into account the structure of input gradients and identifies the updating scheme of optimization. We parameterize the optimizer by the recurrent model and utilize Riemannian operations to ensure that our method is faithful to the geometry of SPD manifolds. Compared with existing SPD optimizers, our optimizer effectively exploits the underlying data distribution and learns a better optimization trajectory in a data-driven manner. Extensive experiments on various computer vision tasks including metric nearness, clustering, and similarity learning demonstrate that our optimizer outperforms existing state-of-the-art methods consistently.

One of the widespread solutions for non-rigid tracking has a nested-loop structure: with Gauss-Newton to minimize a tracking objective in the outer loop, and Preconditioned Conjugate Gradient (PCG) to solve a sparse linear system in the inner loop. In this paper, we employ learnable optimizations to improve tracking robustness and speed up solver convergence. First, we upgrade the tracking objective by integrating an alignment data term on deep features which are learned end-to-end through CNN. The new tracking objective can capture the global deformation which helps Gauss-Newton to jump over local minimum, leading to robust tracking on large non-rigid motions. Second, we bridge the gap between the preconditioning technique and learning method by introducing a ConditionNet which is trained to generate a preconditioner such that PCG can converge within a small number of steps. Experimental results indicate that the proposed learning method converges faster than the original PCG by a large margin.

Many tasks in computer vision are often calibrated and evaluated relative to human perception. In this paper, we propose to directly approximate the perceptual function performed by human observers completing a visual detection task. Specifically, we present a novel methodology for learning to detect image transformations visible to human observers through approximating perceptual thresholds. To do this, we carry out a subjective two-alternative forced-choice study to estimate perceptual thresholds of human observers detecting local exposure shifts in images. We then leverage transformation equivariant representation learning to overcome issues of limited perceptual data. This representation is then used to train a dense convolutional classifier capable of detecting local suprathreshold exposure shifts - a distortion common to image composites. In this context, our model can approximate perceptual thresholds with an average error of 0.1148 exposure stops between empirical and predicted thresholds. It can also be trained to detect a range of different local transformations.

Measuring friction coefficients between cloth and an external body is a longstanding issue in mechanical engineering, never yet addressed with a pure vision-based system. The latter offers the prospect of simpler, less invasive friction measurement protocols compared to traditional ones, and can vastly benefit from recent deep learning advances. Such a novel measurement strategy however proves challenging, as no large labelled dataset for cloth contact exists, and creating one would require thousands of physics workbench measurements with broad coverage of cloth-material pairs. Using synthetic data instead is only possible assuming the availability of a soft-body mechanical simulator with true-to-life friction physics accuracy, yet to be verified. We propose a first vision-based measurement network for friction between cloth and a substrate, using a simple and repeatable video acquisition protocol. We train our network on purely synthetic data generated by a state-of-the-art frictional contact simulator, which we carefully calibrate and validate against real experiments under controlled conditions. We show promising results on a large set of contact pairs between real cloth samples and various kinds of substrates, with 93.6% of all measurements predicted within 0.1 range of standard physics bench measurements.

We describe a method to train a generative model with latent factors that are (approximately) independent and localized. This means that perturbing the latent variables affects only local regions of the synthesized image, corresponding to objects. Unlike other unsupervised generative models, ours enables object-centric manipulation, without requiring object-level annotations, or any form of annotation for that matter. The key to our method is the combination of spatial disentanglement, enforced by a Contextual Information Separation loss, and perceptual cycle-consistency, enforced by a loss that penalizes changes in the image partition in response to perturbations of the latent factors. We test our method's ability to allow independent control of spatial and semantic factors of variability on existing datasets and also introduce two new ones that highlight the limitations of current methods.

We are concerned with a worst-case scenario in model generalization, in the sense that a model aims to perform well on many unseen domains while there is only one single domain available for training. We propose a new method named adversarial domain augmentation to solve this Out-of-Distribution (OOD) generalization problem. The key idea is to leverage adversarial training to create "fictitious" yet "challenging" populations, from which a model can learn to generalize with theoretical guarantees. To facilitate fast and desirable domain augmentation, we cast the model training in a meta-learning scheme and use a Wasserstein Auto-Encoder (WAE) to relax the widely used worst-case constraint. Detailed theoretical analysis is provided to testify our formulation, while extensive experiments on multiple benchmark datasets indicate its superior performance in tackling single domain generalization.

Image cropping aims at improving the framing of an image by removing its extraneous outer areas, which is widely used in the photography and printing industry. In some cases, the aspect ratio of cropping results is specified depending on some conditions. In this paper, we propose a meta-learning (learning to learn) based aspect ratio specified image cropping method called Mars, which can generate cropping results of different expected aspect ratios. In the proposed method, a base model and two meta-learners are obtained during the training stage. Given an aspect ratio in the test stage, a new model with new parameters can be generated from the base model. Specifically, the two meta-learners predict the parameters of the base model based on the given aspect ratio. The learning process of the proposed method is learning how to learn cropping models for different aspect ratio requirements, which is a typical meta-learning process. In the experiments, the proposed method is evaluated on three datasets and outperforms most state-of-the-art methods in terms of accuracy and speed. In addition, both the intermediate and final results show that the proposed model can predict different cropping windows for an image depending on different aspect ratio requirements.

We propose a novel method to use both audio and a low-resolution image to perform extreme face super-resolution (a 16x increase of the input size). When the resolution of the input image is very low (e.g., 8x8 pixels), the loss of information is so dire that important details of the original identity have been lost and audio can aid the recovery of a plausible high-resolution image. In fact, audio carries information about facial attributes, such as gender and age. To combine the aural and visual modalities, we propose a method to first build the latent representations of a face from the lone audio track and then from the lone low-resolution image. We then train a network to fuse these two representations. We show experimentally that audio can assist in recovering attributes such as the gender, the age and the identity, and thus improve the correctness of the high-resolution image reconstruction process. Our procedure does not make use of human annotation and thus can be easily trained with existing video datasets. Moreover, we show that our model builds a factorized representation of images and audio as it allows one to mix low-resolution images and audio from different videos and to generate realistic faces with semantically meaningful combinations.

Synthetic images rendered by graphics engines are a promising source for training deep networks. However, it is challenging to ensure that they can help train a network to perform well on real images, because a graphics-based generation pipeline requires numerous design decisions such as the selection of 3D shapes and the placement of the camera. In this work, we propose a new method that optimizes the generation of 3D training data based on what we call "hybrid gradient". We parametrize the design decisions as a real vector, and combine the approximate gradient and the analytical gradient to obtain the hybrid gradient of the network performance with respect to this vector. We evaluate our approach on the task of estimating surface normal, depth or intrinsic decomposition from a single image. Experiments on standard benchmarks show that our approach can outperform the prior state of the art on optimizing the generation of 3D training data, particularly in terms of computational efficiency.

Few-shot learning is a challenging problem where the goal is to achieve generalization from only few examples. Model-agnostic meta-learning (MAML) tackles the problem by formulating prior knowledge as a common initialization across tasks, which is then used to quickly adapt to unseen tasks. However, forcibly sharing an initialization can lead to conflicts among tasks and the compromised (undesired by tasks) location on optimization landscape, thereby hindering the task adaptation. Further, we observe that the degree of conflict differs among not only tasks but also layers of a neural network. Thus, we propose task-and-layer-wise attenuation on the compromised initialization to reduce its influence. As the attenuation dynamically controls (or selectively forgets) the influence of prior knowledge for a given task and each layer, we name our method as L2F (Learn to Forget). The experimental results demonstrate that the proposed method provides faster adaptation and greatly improves the performance. Furthermore, L2F can be easily applied and improve other state-of-the-art MAML-based frameworks, illustrating its simplicity and generalizability.

Variants of accuracy and precision are the gold-standard by which the computer vision community measures progress of perception algorithms. One reason for the ubiquity of these metrics is that they are largely task-agnostic; we in general seek to detect zero false negatives or positives. The downside of these metrics is that, at worst, they penalize all incorrect detections equally without conditioning on the task or scene, and at best, heuristics need to be chosen to ensure that different mistakes count differently. In this paper, we propose a principled metric for 3D object detection specifically for the task of self-driving. The core idea behind our metric is to isolate the task of object detection and measure the impact the produced detections would induce on the downstream task of driving. Without hand-designing it to, we find that our metric penalizes many of the mistakes that other metrics penalize by design. In addition, our metric downweighs detections based on additional factors such as distance from a detection to the ego car and the speed of the detection in intuitive ways that other detection metrics do not. For human evaluation, we generate scenes in which standard metrics and our metric disagree and find that humans side with our metric 79% of the time. Our project page including an evaluation server can be found at https://nv-tlabs.github.io/detection-relevance.