ICCV 2019 Papers

We present two novel solutions for multi-view 3D human pose estimation based on new learnable triangulation methods that combine 3D information from multiple 2D views. The first (baseline) solution is a basic differentiable algebraic triangulation with an addition of confidence weights estimated from the input images. The second, more complex, solution is based on volumetric aggregation of 2D feature maps from the 2D backbone followed by refinement via 3D convolutions that produce final 3D joint heatmaps. Crucially, both of the approaches are end-to-end differentiable, which allows us to directly optimize the target metric. We demonstrate transferability of the solutions across datasets and considerably improve the multi-view state of the art on the Human3.6M dataset.

Domain adaptation enables the learner to safely generalize into novel environments by mitigating domain shifts across distributions. Previous works may not effectively uncover the underlying reasons that would lead to the drastic model degradation on the target task. In this paper, we empirically reveal that the erratic discrimination of the target domain mainly stems from its much smaller feature norms with respect to that of the source domain. To this end, we propose a novel parameter-free Adaptive Feature Norm approach. We demonstrate that progressively adapting the feature norms of the two domains to a large range of values can result in significant transfer gains, implying that those task-specific features with larger norms are more transferable. Our method successfully unifies the computation of both standard and partial domain adaptation with more robustness against the negative transfer issue. Without bells and whistles but a few lines of code, our method substantially lifts the performance on the target task and exceeds state-of-the-arts by a large margin (11.5% on Office-Home and 17.1% on VisDA2017). We hope our simple yet effective approach will shed some light on the future research of transfer learning. Code is available at https://github.com/jihanyang/AFN.

Distillation-based learning boosts the performance of the miniaturized neural network based on the hypothesis that the representation of a teacher model can be used as structured and relatively weak supervision, and thus would be easily learned by a miniaturized model. However, we find that the representation of a converged heavy model is still a strong constraint for training a small student model, which leads to a higher lower bound of congruence loss. In this work, we consider the knowledge distillation from the perspective of curriculum learning by teacher's routing. Instead of supervising the student model with a converged teacher model, we supervised it with some anchor points selected from the route in parameter space that the teacher model passed by, as we called route constrained optimization (RCO). We experimentally demonstrate this simple operation greatly reduces the lower bound of congruence loss for knowledge distillation, hint and mimicking learning. On close-set classification tasks like CIFAR and ImageNet, RCO improves knowledge distillation by 2.14% and 1.5% respectively. For the sake of evaluating the generalization, we also test RCO on the open-set face recognition task MegaFace. RCO achieves 84.3% accuracy on one-to-million task with only 0.8 M parameters, which push the SOTA by a large margin.

Point cloud is an important type of 3D representation. However, directly applying convolutions on point clouds is challenging due to the sparse, irregular and unordered data structure. In this paper, we propose a novel Interpolated Convolution operation, InterpConv, to tackle the point cloud feature learning and understanding problem. The key idea is to utilize a set of discrete kernel weights and interpolate point features to neighboring kernel-weight coordinates by an interpolation function for convolution. A normalization term is introduced to handle neighborhoods of different sparsity levels. Our InterpConv is shown to be permutation and sparsity invariant, and can directly handle irregular inputs. We further design Interpolated Convolutional Neural Networks (InterpCNNs) based on InterpConv layers to handle point cloud recognition tasks including shape classification, object part segmentation and indoor scene semantic parsing. Experiments show that the networks can capture both fine-grained local structures and global shape context information effectively. The proposed approach achieves state-of-the-art performance on public benchmarks including ModelNet40, ShapeNet Parts and S3DIS.

Generative Adversarial Networks (GANs) typically learn a distribution of images in a large image dataset, and are then able to generate new images from this distribution. However, each natural image has its own internal statistics, captured by its unique distribution of patches. In this paper we propose an "Internal GAN" (InGAN) -- an image-specific GAN -- which trains on a single input image and learns its internal distribution of patches. It is then able to synthesize a plethora of new natural images of significantly different sizes, shapes and aspect-ratios - all with the same internal patch-distribution (same "DNA") as the input image. In particular, despite large changes in global size/shape of the image, all elements inside the image maintain their local size/shape. InGAN is fully unsupervised, requiring no additional data other than the input image itself. Once trained on the input image, it can remap the input to any size or shape in a single feedforward pass, while preserving the same internal patch distribution. InGAN provides a unified framework for a variety of tasks, bridging the gap between textures and natural images.

We propose an efficient algorithm to embed a given image into the latent space of StyleGAN. This embedding enables semantic image editing operations that can be applied to existing photographs. Taking the StyleGAN trained on the FFHD dataset as an example, we show results for image morphing, style transfer, and expression transfer. Studying the results of the embedding algorithm provides valuable insights into the structure of the StyleGAN latent space. We propose a set of experiments to test what class of images can be embedded, how they are embedded, what latent space is suitable for embedding, and if the embedding is semantically meaningful.

We present Habitat, a platform for research in embodied artificial intelligence (AI). Habitat enables training embodied agents (virtual robots) in highly efficient photorealistic 3D simulation. Specifically, Habitat consists of: (i) Habitat-Sim: a flexible, high-performance 3D simulator with configurable agents, sensors, and generic 3D dataset handling. Habitat-Sim is fast -- when rendering a scene from Matterport3D, it achieves several thousand frames per second (fps) running single-threaded, and can reach over 10,000 fps multi-process on a single GPU. (ii) Habitat-API: a modular high-level library for end-to-end development of embodied AI algorithms -- defining tasks (e.g., navigation, instruction following, question answering), configuring, training, and benchmarking embodied agents. These large-scale engineering contributions enable us to answer scientific questions requiring experiments that were till now impracticable or 'merely' impractical. Specifically, in the context of point-goal navigation: (1) we revisit the comparison between learning and SLAM approaches from two recent works and find evidence for the opposite conclusion -- that learning outperforms SLAM if scaled to an order of magnitude more experience than previous investigations, and (2) we conduct the first cross-dataset generalization experiments train, test x Matterport3D, Gibson for multiple sensors blind, RGB, RGBD, D and find that only agents with depth (D) sensors generalize across datasets. We hope that our open-source platform and these findings will advance research in embodied AI.

The fully-convolutional siamese network based on template matching has shown great potentials in visual tracking. During testing, the template is fixed with the initial target feature and the performance totally relies on the general matching ability of the siamese network. However, this manner cannot capture the temporal variations of targets or background clutter. In this work, we propose a novel gradient-guided network to exploit the discriminative information in gradients and update the template in the siamese network through feed-forward and backward operations. To be specific, the algorithm can utilize the information from the gradient to update the template in the current frame. In addition, a template generalization training method is proposed to better use gradient information and avoid overfitting. To our knowledge, this work is the first attempt to exploit the information in the gradient for template update in siamese-based trackers. Extensive experiments on recent benchmarks demonstrate that our method achieves better performance than other state-of-the-art trackers.

Multi-view action recognition targets to integrate complementary information from different views to improve classification performance. It is a challenging task due to the distinct gap between heterogeneous feature domains. Moreover, most existing methods neglect to consider the incomplete multi-view data, which limits their potential compatibility in real-world applications. In this work, we propose a Generative Multi-View Action Recognition (GMVAR) framework to address the challenges above. The adversarial generative network is leveraged to generate one view conditioning on the other view, which fully explores the latent connections in both intra-view and cross-view aspects. Our approach enhances the model robustness by employing adversarial training, and naturally handles the incomplete view case by imputing the missing data. Moreover, an effective View Correlation Discovery Network (VCDN) is proposed to further fuse the multi-view information in a higher-level label space. Extensive experiments demonstrate the effectiveness of our proposed approach by comparing with state-of-the-art algorithms.

This paper explores object detection in the small data regime, where only a limited number of annotated bounding boxes are available due to data rarity and annotation expense. This is a common challenge today with machine learning being applied to many new tasks where obtaining training data is more challenging, e.g. in medical images with rare diseases that doctors sometimes only see once in their life-time. In this work we explore this problem from a generative modeling perspective by learning to generate new images with associated bounding boxes, and using these for training an object detector. We show that simply training previously proposed generative models does not yield satisfactory performance due to them optimizing for image realism rather than object detection accuracy. To this end we develop a new model with a novel unrolling mechanism that jointly optimizes the generative model and a detector such that the generated images improve the performance of the detector. We show this method outperforms the state of the art on two challenging datasets, disease detection and small data pedestrian detection, improving the average precision on NIH Chest X-ray by a relative 20% and localization accuracy by a relative 50%.

We present an imaging framework which converts three images from a gated camera into high-resolution depth maps with depth accuracy comparable to pulsed lidar measurements. Existing scanning lidar systems achieve low spatial resolution at large ranges due to mechanically-limited angular sampling rates, restricting scene understanding tasks to close-range clusters with dense sampling. Moreover, today's pulsed lidar scanners suffer from high cost, power consumption, large form-factors, and they fail in the presence of strong backscatter. We depart from point scanning and demonstrate that it is possible to turn a low-cost CMOS gated imager into a dense depth camera with at least 80m range - by learning depth from three gated images. The proposed architecture exploits semantic context across gated slices, and is trained on a synthetic discriminator loss without the need of dense depth labels. The proposed replacement for scanning lidar systems is real-time, handles back-scatter and provides dense depth at long ranges. We validate our approach in simulation and on real-world data acquired over 4,000km driving in northern Europe. Data and code are available at https://github.com/gruberto/Gated2Depth.

Traditional intrinsic image decomposition focuses on decomposing images into reflectance and shading, leaving surfaces normals and lighting entangled in shading. In this work, we propose a Global-Local Spherical Harmonics (GLoSH) lighting model to improve the lighting component, and jointly predict reflectance and surface normals. The global SH models the holistic lighting while local SH account for the spatial variation of lighting. Also, a novel non-negative lighting constraint is proposed to encourage the estimated SH to be physically meaningful. To seamlessly reflect the GLoSH model, we design a coarse-to-fine network structure. The coarse network predicts global SH, reflectance and normals, and the fine network predicts their local residuals. Lacking labels for reflectance and lighting, we apply synthetic data for model pre-training and fine-tune the model with real data in a self-supervised way. Compared to the state-of-the-art methods only targeting normals or reflectance and shading, our method recovers all components and achieves consistently better results on three real datasets, IIW, SAW and NYUv2.

Text present in images are not merely strings, they provide useful cues about the image. Despite their utility in better image understanding, scene texts are not used in traditional visual question answering (VQA) models. In this work, we present a VQA model which can read scene texts and perform reasoning on a knowledge graph to arrive at an accurate answer. Our proposed model has three mutually interacting modules: i. proposal module to get word and visual content proposals from the image, ii. fusion module to fuse these proposals, question and knowledge base to mine relevant facts, and represent these facts as multi-relational graph, iii. reasoning module to perform a novel gated graph neural network based reasoning on this graph. The performance of our knowledge-enabled VQA model is evaluated on our newly introduced dataset, viz. text-KVQA. To the best of our knowledge, this is the first dataset which identifies the need for bridging text recognition with knowledge graph based reasoning. Through extensive experiments, we show that our proposed method outperforms traditional VQA as well as question-answering over knowledge base-based methods on text-KVQA.

We present a generative image inpainting system to complete images with free-form mask and guidance. The system is based on gated convolutions learned from millions of images without additional labelling efforts. The proposed gated convolution solves the issue of vanilla convolution that treats all input pixels as valid ones, generalizes partial convolution by providing a learnable dynamic feature selection mechanism for each channel at each spatial location across all layers. Moreover, as free-form masks may appear anywhere in images with any shape, global and local GANs designed for a single rectangular mask are not applicable. Thus, we also present a patch-based GAN loss, named SN-PatchGAN, by applying spectral-normalized discriminator on dense image patches. SN-PatchGAN is simple in formulation, fast and stable in training. Results on automatic image inpainting and user-guided extension demonstrate that our system generates higher-quality and more flexible results than previous methods. Our system helps user quickly remove distracting objects, modify image layouts, clear watermarks and edit faces. Code, demo and models are available at: https://github.com/JiahuiYu/generative_inpainting.

Visual compatibility is critical for fashion analysis, yet is missing in existing fashion image synthesis systems. In this paper, we propose to explicitly model visual compatibility through fashion image inpainting. We present Fashion Inpainting Networks (FiNet), a two-stage image-to-image generation framework that is able to perform compatible and diverse inpainting. Disentangling the generation of shape and appearance to ensure photorealistic results, our framework consists of a shape generation network and an appearance generation network. More importantly, for each generation network, we introduce two encoders interacting with one another to learn latent codes in a shared compatibility space. The latent representations are jointly optimized with the corresponding generation network to condition the synthesis process, encouraging a diverse set of generated results that are visually compatible with existing fashion garments. In addition, our framework is readily extended to clothing reconstruction and fashion transfer. Extensive experiments on fashion synthesis quantitatively and qualitatively demonstrate the effectiveness of our method.

Several recent works have shown how highly realistic human head images can be obtained by training convolutional neural networks to generate them. In order to create a personalized talking head model, these works require training on a large dataset of images of a single person. However, in many practical scenarios, such personalized talking head models need to be learned from a few image views of a person, potentially even a single image. Here, we present a system with such few-shot capability. It performs lengthy meta-learning on a large dataset of videos, and after that is able to frame few- and one-shot learning of neural talking head models of previously unseen people as adversarial training problems with high capacity generators and discriminators. Crucially, the system is able to initialize the parameters of both the generator and the discriminator in a person-specific way, so that training can be based on just a few images and done quickly, despite the need to tune tens of millions of parameters. We show that such an approach is able to learn highly realistic and personalized talking head models of new people and even portrait paintings.

Inter-personal anatomical differences limit the accuracy of person-independent gaze estimation networks. Yet there is a need to lower gaze errors further to enable applications requiring higher quality. Further gains can be achieved by personalizing gaze networks, ideally with few calibration samples. However, over-parameterized neural networks are not amenable to learning from few examples as they can quickly over-fit. We embrace these challenges and propose a novel framework for Few-shot Adaptive GaZE Estimation (Faze) for learning person-specific gaze networks with very few (<= 9) calibration samples. Faze learns a rotation-aware latent representation of gaze via a disentangling encoder-decoder architecture along with a highly adaptable gaze estimator trained using meta-learning. It is capable of adapting to any new person to yield significant performance gains with as few as 3 samples, yielding state-of-the-art performance of 3.18-deg on GazeCapture, a 19% improvement over prior art. We open-source our code at https://github.com/NVlabs/few_shot_gaze

Matching clothing images from customers and online shopping stores has rich applications in E-commerce. Existing algorithms encoded an image as a global feature vector and performed retrieval with the global representation. However, discriminative local information on clothes are submerged in this global representation, resulting in sub-optimal performance. To address this issue, we propose a novel Graph Reasoning Network (GRNet) on a Similarity Pyramid, which learns similarities between a query and a gallery cloth by using both global and local representations in multiple scales. The similarity pyramid is represented by a Graph of similarity, where nodes represent similarities between clothing components at different scales, and the final matching score is obtained by message passing along edges. In GRNet, graph reasoning is solved by training a graph convolutional network, enabling to align salient clothing components to improve clothing retrieval. To facilitate future researches, we introduce a new benchmark FindFashion, containing rich annotations of bounding boxes, views, occlusions, and cropping. Extensive experiments show that GRNet obtains new state-of-the-art results on two challenging benchmarks, e.g. pushing the top-1, top-20, and top-50 accuracies on DeepFashion to 26%, 64%, and 75% (i.e. 4%, 10%, and 10% absolute improvements), outperforming competitors with large margins. On FindFashion, GRNet achieves considerable improvements on all empirical settings.

Existing face deblurring methods only consider single frames and do not account for facial structure and identity information. These methods struggle to deblur face videos that exhibit significant pose variations and misalignment. In this paper we propose a novel face video deblurring network capitalizing on 3D facial priors. The model consists of two main branches: i) a face video deblurring sub-network based on an encoder-decoder architecture, and ii) a 3D face reconstruction and rendering branch for predicting 3D priors of salient facial structures and identity knowledge. These structures encourage the deblurring branch to generate sharp faces with detailed structures. Our method not only uses low-level information (i.e., image intensity), but also middle-level information (i.e., 3D facial structure) and high-level knowledge (i.e., identity content) to further explore spatial constraints of facial components from blurry face frames. Extensive experimental results demonstrate that the proposed algorithm performs favorably against the state-of-the-art methods.