With the capacity of modeling long-range dependencies in sequential data, transformers have shown remarkable performances in a variety of generative tasks such as image, audio, and text generation. Yet, taming them in generating less structured and voluminous data formats such as high-resolution point clouds have seldom been explored due to ambiguous sequentialization processes and infeasible computation burden. In this paper, we aim to further exploit the power of transformers and employ them for the task of 3D point cloud generation. The key idea is to decompose point clouds of one category into semantically aligned sequences of shape compositions, via a learned canonical space. These shape compositions can then be quantized and used to learn a context-rich composition codebook for point cloud generation. Experimental results on point cloud reconstruction and unconditional generation show that our model performs favorably against state-of-the-art approaches. Furthermore, our model can be easily extended to multi-modal shape completion as an application for conditional shape generation.
This paper aims at representing animatable photo-realistic humans under novel views and poses. Recent work has shown significant progress with dynamic scenes by exploring shared canonical neural radiance fields. However learning a user-controlled model for novel poses remains a challenging task. To tackle this problem, we introduce a novel method to integrate observations across frames and encode the appearance at each individual frame by utilizing the human pose that models the body shape and point clouds which cover partial part of the human as the input. Specifically, our method simultaneously learns a shared set of latent codes anchored to the human pose among frames, and learns an appearance-dependent code anchored to incomplete point clouds generated by monocular RGB-D at each frame. A human pose-based code models the shape of the performer whereas a point cloud based code predicts details and reasons about missing structures at the unseen poses. To further recover non-visible regions in query frames, we utilize a temporal transformer to integrate features of points in query frames and tracked body points from automatically-selected key frames. Experiments on various sequences of humans in motion show that our method significantly outperforms existing works under unseen poses and novel views given monocular RGB-D videos as input.
While recent face anti-spoofing methods perform well under the intra-domain setups, an effective approach needs to account for much larger appearance variations of images acquired in complex scenes with different sensors for robust performance. In this paper, we present adaptive vision transformers (ViT) for robust cross-domain face anti-spoofing. Specifically, we adopt ViT as a backbone to exploit its strength to account for long-range dependencies among pixels. We further introduce the ensemble adapters module and feature-wise transformation layers in the ViT to adapt to different domains for robust performance with a few samples. Experiments on several benchmark datasets show that the proposed models achieve both robust and competitive performance against the state-of-the-art methods.
In this paper, we investigate the application of Vehicle-to-Everything (V2X) communication to improve the perception performance of autonomous vehicles. We present a robust cooperative perception framework with V2X communication using a novel vision Transformer. Specifically, we build a holistic attention model, namely V2X-ViT, to effectively fuse information across on-road agents (i.e., vehicles and infrastructure). V2X-ViT consists of alternating layers of heterogeneous multi-agent self-attention and multi-scale window self-attention, which captures inter-agent interaction and per-agent spatial relationships. These key modules are designed in a unified Transformer architecture to handle common V2X challenges, including asynchronous information sharing, pose errors, and heterogeneity of V2X components. To validate our approach, we create a large-scale V2X perception dataset using CARLA and OpenCDA. Extensive experimental results demonstrate that V2X-ViT sets new state-of-the-art performance for 3D object detection and achieves robust performance even under harsh, noisy environments. The dataset, source code, and trained models will be open-sourced.
Image deblurring is a classic problem in low-level computer vision, which aims to recover a sharp image from a blurred input image. Recent advances in deep learning have led to significant progress in solving this problem, and a large number of deblurring networks have been proposed. This paper presents a comprehensive and timely survey of recently published deep-learning based image deblurring approaches, aiming to serve the community as a useful literature review. We start by discussing common causes of image blur, introduce benchmark datasets and performance metrics, and summarize different problem formulations. Next we present a taxonomy of methods using convolutional neural networks (CNN) based on architecture, loss function, and application, offering a detailed review and comparison. In addition, we discuss some domain-specific deblurring applications including face images, text, and stereo image pairs. We conclude by discussing key challenges and future research directions.
In this paper, we explore the possibility of building a unified foundation model that can be adapted to both vision-only and text-only tasks. Starting from BERT and ViT, we design a unified transformer consisting of modality-specific tokenizers, a shared transformer encoder, and task-specific output heads. To efficiently pre-train the proposed model jointly on unpaired images and text, we propose two novel techniques: (i) We employ the separately-trained BERT and ViT models as teachers and apply knowledge distillation to provide additional, accurate supervision signals for the joint training; (ii) We propose a novel gradient masking strategy to balance the parameter updates from the image and text pre-training losses. We evaluate the jointly pre-trained transformer by fine-tuning it on image classification tasks and natural language understanding tasks, respectively. The experiments show that the resultant unified foundation transformer works surprisingly well on both the vision-only and text-only tasks, and the proposed knowledge distillation and gradient masking strategy can effectively lift the performance to approach the level of separately-trained models.
Along with the rapid progress of visual tracking, existing benchmarks become less informative due to redundancy of samples and weak discrimination between current trackers, making evaluations on all datasets extremely time-consuming. Thus, a small and informative benchmark, which covers all typical challenging scenarios to facilitate assessing the tracker performance, is of great interest. In this work, we develop a principled way to construct a small and informative tracking benchmark (ITB) with 7% out of 1.2 M frames of existing and newly collected datasets, which enables efficient evaluation while ensuring effectiveness. Specifically, we first design a quality assessment mechanism to select the most informative sequences from existing benchmarks taking into account 1) challenging level, 2) discriminative strength, 3) and density of appearance variations. Furthermore, we collect additional sequences to ensure the diversity and balance of tracking scenarios, leading to a total of 20 sequences for each scenario. By analyzing the results of 15 state-of-the-art trackers re-trained on the same data, we determine the effective methods for robust tracking under each scenario and demonstrate new challenges for future research direction in this field.
A modern self-supervised learning algorithm typically enforces persistency of the representations of an instance across views. While being very effective on learning holistic image and video representations, such an approach becomes sub-optimal for learning spatio-temporally fine-grained features in videos, where scenes and instances evolve through space and time. In this paper, we present the Contextualized Spatio-Temporal Contrastive Learning (ConST-CL) framework to effectively learn spatio-temporally fine-grained representations using self-supervision. We first design a region-based self-supervised pretext task which requires the model to learn to transform instance representations from one view to another guided by context features. Further, we introduce a simple network design that effectively reconciles the simultaneous learning process of both holistic and local representations. We evaluate our learned representations on a variety of downstream tasks and ConST-CL achieves state-of-the-art results on four datasets. For spatio-temporal action localization, ConST-CL achieves 39.4% mAP with ground-truth boxes and 30.5% mAP with detected boxes on the AVA-Kinetics validation set. For object tracking, ConST-CL achieves 78.1% precision and 55.2% success scores on OTB2015. Furthermore, ConST-CL achieves 94.8% and 71.9% top-1 fine-tuning accuracy on video action recognition datasets, UCF101 and HMDB51 respectively. We plan to release our code and models to the public.
This work presents a self-supervised learning framework named TeG to explore Temporal Granularity in learning video representations. In TeG, we sample a long clip from a video and a short clip that lies inside the long clip. We then extract their dense temporal embeddings. The training objective consists of two parts: a fine-grained temporal learning objective to maximize the similarity between corresponding temporal embeddings in the short clip and the long clip, and a persistent temporal learning objective to pull together global embeddings of the two clips. Our study reveals the impact of temporal granularity with three major findings. 1) Different video tasks may require features of different temporal granularities. 2) Intriguingly, some tasks that are widely considered to require temporal awareness can actually be well addressed by temporally persistent features. 3) The flexibility of TeG gives rise to state-of-the-art results on 8 video benchmarks, outperforming supervised pre-training in most cases.
Blur artifacts can seriously degrade the visual quality of images, and numerous deblurring methods have been proposed for specific scenarios. However, in most real-world images, blur is caused by different factors, e.g., motion and defocus. In this paper, we address how different deblurring methods perform on general types of blur. For in-depth performance evaluation, we construct a new large-scale multi-cause image deblurring dataset called (MC-Blur) including real-world and synthesized blurry images with mixed factors of blurs. The images in the proposed MC-Blur dataset are collected using different techniques: convolving Ultra-High-Definition (UHD) sharp images with large kernels, averaging sharp images captured by a 1000 fps high-speed camera, adding defocus to images, and real-world blurred images captured by various camera models. These results provide a comprehensive overview of the advantages and limitations of current deblurring methods. Further, we propose a new baseline model, level-attention deblurring network, to adapt to multiple causes of blurs. By including different weights of attention to the different levels of features, the proposed network derives more powerful features with larger weights assigned to more important levels, thereby enhancing the feature representation. Extensive experimental results on the new dataset demonstrate the effectiveness of the proposed model for the multi-cause blur scenarios.