Conditional Directed Graph Convolution for 3D Human Pose Estimation

Graph convolutional networks have significantly improved 3D human pose estimation by representing the human skeleton as an undirected graph. However, this representation fails to reflect the articulated characteristic of human skeletons as the hierarchical orders among the joints are not explicitly presented. In this paper, we propose to represent the human skeleton as a directed graph with the joints as nodes and bones as edges that are directed from parent joints to child joints. By so doing, the directions of edges can explicitly reflect the hierarchical relationships among the nodes. Based on this representation, we adopt the spatial-temporal directed graph convolution (ST-DGConv) to extract features from 2D poses represented in a temporal sequence of directed graphs. We further propose a spatial-temporal conditional directed graph convolution (ST-CondDGConv) to leverage varying non-local dependence for different poses by conditioning the graph topology on input poses. Altogether, we form a U-shaped network with ST-DGConv and ST-CondDGConv layers, named U-shaped Conditional Directed Graph Convolutional Network (U-CondDGCN), for 3D human pose estimation from monocular videos. To evaluate the effectiveness of our U-CondDGCN, we conducted extensive experiments on two challenging large-scale benchmarks: Human3.6M and MPI-INF-3DHP. Both quantitative and qualitative results show that our method achieves top performance. Also, ablation studies show that directed graphs can better exploit the hierarchy of articulated human skeletons than undirected graphs, and the conditional connections can yield adaptive graph topologies for different kinds of poses.

SynLiDAR: Learning From Synthetic LiDAR Sequential Point Cloud for Semantic Segmentation

Transfer learning from synthetic to real data has been proved an effective way of mitigating data annotation constraints in various computer vision tasks. However, the developments focused on 2D images but lag far behind for 3D point clouds due to the lack of large-scale high-quality synthetic point cloud data and effective transfer methods. We address this issue by collecting SynLiDAR, a synthetic LiDAR point cloud dataset that contains large-scale point-wise annotated point cloud with accurate geometric shapes and comprehensive semantic classes, and designing PCT-Net, a point cloud translation network that aims to narrow down the gap with real-world point cloud data. For SynLiDAR, we leverage graphic tools and professionals who construct multiple realistic virtual environments with rich scene types and layouts where annotated LiDAR points can be generated automatically. On top of that, PCT-Net disentangles synthetic-to-real gaps into an appearance component and a sparsity component and translates SynLiDAR by aligning the two components with real-world data separately. Extensive experiments over multiple data augmentation and semi-supervised semantic segmentation tasks show very positive outcomes - including SynLiDAR can either train better models or reduce real-world annotated data without sacrificing performance, and PCT-Net translated data further improve model performance consistently.

FBC-GAN: Diverse and Flexible Image Synthesis via Foreground-Background Composition

Generative Adversarial Networks (GANs) have become the de-facto standard in image synthesis. However, without considering the foreground-background decomposition, existing GANs tend to capture excessive content correlation between foreground and background, thus constraining the diversity in image generation. This paper presents a novel Foreground-Background Composition GAN (FBC-GAN) that performs image generation by generating foreground objects and background scenes concurrently and independently, followed by composing them with style and geometrical consistency. With this explicit design, FBC-GAN can generate images with foregrounds and backgrounds that are mutually independent in contents, thus lifting the undesirably learned content correlation constraint and achieving superior diversity. It also provides excellent flexibility by allowing the same foreground object with different background scenes, the same background scene with varying foreground objects, or the same foreground object and background scene with different object positions, sizes and poses. It can compose foreground objects and background scenes sampled from different datasets as well. Extensive experiments over multiple datasets show that FBC-GAN achieves competitive visual realism and superior diversity as compared with state-of-the-art methods.

Bi-level Feature Alignment for Versatile Image Translation and Manipulation

Generative adversarial networks (GANs) have achieved great success in image translation and manipulation. However, high-fidelity image generation with faithful style control remains a grand challenge in computer vision. This paper presents a versatile image translation and manipulation framework that achieves accurate semantic and style guidance in image generation by explicitly building a correspondence. To handle the quadratic complexity incurred by building the dense correspondences, we introduce a bi-level feature alignment strategy that adopts a top-$k$ operation to rank block-wise features followed by dense attention between block features which reduces memory cost substantially. As the top-$k$ operation involves index swapping which precludes the gradient propagation, we propose to approximate the non-differentiable top-$k$ operation with a regularized earth mover's problem so that its gradient can be effectively back-propagated. In addition, we design a novel semantic position encoding mechanism that builds up coordinate for each individual semantic region to preserve texture structures while building correspondences. Further, we design a novel confidence feature injection module which mitigates mismatch problem by fusing features adaptively according to the reliability of built correspondences. Extensive experiments show that our method achieves superior performance qualitatively and quantitatively as compared with the state-of-the-art. The code is available at \href{https://github.com/fnzhan/RABIT}{https://github.com/fnzhan/RABIT}.

Blind Image Super-Resolution via Contrastive Representation Learning

Image super-resolution (SR) research has witnessed impressive progress thanks to the advance of convolutional neural networks (CNNs) in recent years. However, most existing SR methods are non-blind and assume that degradation has a single fixed and known distribution (e.g., bicubic) which struggle while handling degradation in real-world data that usually follows a multi-modal, spatially variant, and unknown distribution. The recent blind SR studies address this issue via degradation estimation, but they do not generalize well to multi-source degradation and cannot handle spatially variant degradation. We design CRL-SR, a contrastive representation learning network that focuses on blind SR of images with multi-modal and spatially variant distributions. CRL-SR addresses the blind SR challenges from two perspectives. The first is contrastive decoupling encoding which introduces contrastive learning to extract resolution-invariant embedding and discard resolution-variant embedding under the guidance of a bidirectional contrastive loss. The second is contrastive feature refinement which generates lost or corrupted high-frequency details under the guidance of a conditional contrastive loss. Extensive experiments on synthetic datasets and real images show that the proposed CRL-SR can handle multi-modal and spatially variant degradation effectively under blind settings and it also outperforms state-of-the-art SR methods qualitatively and quantitatively.

Sparse Needlets for Lighting Estimation with Spherical Transport Loss

Accurate lighting estimation is challenging yet critical to many computer vision and computer graphics tasks such as high-dynamic-range (HDR) relighting. Existing approaches model lighting in either frequency domain or spatial domain which is insufficient to represent the complex lighting conditions in scenes and tends to produce inaccurate estimation. This paper presents NeedleLight, a new lighting estimation model that represents illumination with needlets and allows lighting estimation in both frequency domain and spatial domain jointly. An optimal thresholding function is designed to achieve sparse needlets which trims redundant lighting parameters and demonstrates superior localization properties for illumination representation. In addition, a novel spherical transport loss is designed based on optimal transport theory which guides to regress lighting representation parameters with consideration of the spatial information. Furthermore, we propose a new metric that is concise yet effective by directly evaluating the estimated illumination maps rather than rendered images. Extensive experiments show that NeedleLight achieves superior lighting estimation consistently across multiple evaluation metrics as compared with state-of-the-art methods.

Unbalanced Feature Transport for Exemplar-based Image Translation

Despite the great success of GANs in images translation with different conditioned inputs such as semantic segmentation and edge maps, generating high-fidelity realistic images with reference styles remains a grand challenge in conditional image-to-image translation. This paper presents a general image translation framework that incorporates optimal transport for feature alignment between conditional inputs and style exemplars in image translation. The introduction of optimal transport mitigates the constraint of many-to-one feature matching significantly while building up accurate semantic correspondences between conditional inputs and exemplars. We design a novel unbalanced optimal transport to address the transport between features with deviational distributions which exists widely between conditional inputs and exemplars. In addition, we design a semantic-activation normalization scheme that injects style features of exemplars into the image translation process successfully. Extensive experiments over multiple image translation tasks show that our method achieves superior image translation qualitatively and quantitatively as compared with the state-of-the-art.

Diverse Image Inpainting with Bidirectional and Autoregressive Transformers

Image inpainting is an underdetermined inverse problem, it naturally allows diverse contents that fill up the missing or corrupted regions reasonably and realistically. Prevalent approaches using convolutional neural networks (CNNs) can synthesize visually pleasant contents, but CNNs suffer from limited perception fields for capturing global features. With image-level attention, transformers enable to model long-range dependencies and generate diverse contents with autoregressive modeling of pixel-sequence distributions. However, the unidirectional attention in transformers is suboptimal as corrupted regions can have arbitrary shapes with contexts from arbitrary directions. We propose BAT-Fill, an image inpainting framework with a novel bidirectional autoregressive transformer (BAT) that models deep bidirectional contexts for autoregressive generation of diverse inpainting contents. BAT-Fill inherits the merits of transformers and CNNs in a two-stage manner, which allows to generate high-resolution contents without being constrained by the quadratic complexity of attention in transformers. Specifically, it first generates pluralistic image structures of low resolution by adapting transformers and then synthesizes realistic texture details of high resolutions with a CNN-based up-sampling network. Extensive experiments over multiple datasets show that BAT-Fill achieves superior diversity and fidelity in image inpainting qualitatively and quantitatively.

Deep Monocular 3D Human Pose Estimation via Cascaded Dimension-Lifting

The 3D pose estimation from a single image is a challenging problem due to depth ambiguity. One type of the previous methods lifts 2D joints, obtained by resorting to external 2D pose detectors, to the 3D space. However, this type of approaches discards the contextual information of images which are strong cues for 3D pose estimation. Meanwhile, some other methods predict the joints directly from monocular images but adopt a 2.5D output representation $P^{2.5D} = (u,v,z^{r}) $ where both $u$ and $v$ are in the image space but $z^{r}$ in root-relative 3D space. Thus, the ground-truth information (e.g., the depth of root joint from the camera) is normally utilized to transform the 2.5D output to the 3D space, which limits the applicability in practice. In this work, we propose a novel end-to-end framework that not only exploits the contextual information but also produces the output directly in the 3D space via cascaded dimension-lifting. Specifically, we decompose the task of lifting pose from 2D image space to 3D spatial space into several sequential sub-tasks, 1) kinematic skeletons \& individual joints estimation in 2D space, 2) root-relative depth estimation, and 3) lifting to the 3D space, each of which employs direct supervisions and contextual image features to guide the learning process. Extensive experiments show that the proposed framework achieves state-of-the-art performance on two widely used 3D human pose datasets (Human3.6M, MuPoTS-3D).