Estimating the 2D human poses in each view is typically the first step in calibrated multi-view 3D pose estimation. But the performance of 2D pose detectors suffers from challenging situations such as occlusions and oblique viewing angles. To address these challenges, previous works derive point-to-point correspondences between different views from epipolar geometry and utilize the correspondences to merge prediction heatmaps or feature representations. Instead of post-prediction merge/calibration, here we introduce a transformer framework for multi-view 3D pose estimation, aiming at directly improving individual 2D predictors by integrating information from different views. Inspired by previous multi-modal transformers, we design a unified transformer architecture, named TransFusion, to fuse cues from both current views and neighboring views. Moreover, we propose the concept of epipolar field to encode 3D positional information into the transformer model. The 3D position encoding guided by the epipolar field provides an efficient way of encoding correspondences between pixels of different views. Experiments on Human 3.6M and Ski-Pose show that our method is more efficient and has consistent improvements compared to other fusion methods. Specifically, we achieve 25.8 mm MPJPE on Human 3.6M with only 5M parameters on 256 x 256 resolution.
Hong Kong University of Science and Technology（香港科技大学）；Huawei（华为）；City University of Hong Kong（香港城市大学）
LiDAR and camera are two important sensors for 3D object detection in autonomous driving. Despite the increasing popularity of sensor fusion in this field, the robustness against inferior image conditions, e.g., bad illumination and sensor misalignment, is under-explored. Existing fusion methods are easily affected by such conditions, mainly due to a hard association of LiDAR points and image pixels, established by calibration matrices. We propose TransFusion, a robust solution to LiDAR-camera fusion with a soft-association mechanism to handle inferior image conditions. Specifically, our Trans…
Combining information from multi-view images is crucial to improve the performance and robustness of automated methods for disease diagnosis. However, due to the non-alignment characteristics of multi-view images, building correlation and data fusion across views largely remain an open problem. In this study, we present TransFusion, a Transformer-based architecture to merge divergent multi-view imaging information using convolutional layers and powerful attention mechanisms. In particular, the Divergent Fusion Attention (DiFA) module is proposed for rich cross-view context modeling and semanti…
Peking University（北京大学）；Deepwise AI Lab（深睿医疗人工智能实验室）；Microsoft Research Asia（微软亚洲研究院）
Cross view feature fusion is the key to address the occlusion problem in human pose estimation. The current fusion methods need to train a separate model for every pair of cameras making them difficult to scale. In this work, we introduce MetaFuse, a pre-trained fusion model learned from a large number of cameras in the Panoptic dataset. The model can be efficiently adapted or finetuned for a new pair of cameras using a small number of labeled images. The strong adaptation power of MetaFuse is due in large part to the proposed factorization of the original fusion model into two parts (1) a gen…
University of Science and Technology of China（中国科学技术大学）；Microsoft Research Asia（微软亚洲研究院）；TuSimple（图森未来）
We present an approach to recover absolute 3D human poses from multi-view images by incorporating multi-view geometric priors in our model. It consists of two separate steps: (1) estimating the 2D poses in multi-view images and (2) recovering the 3D poses from the multi-view 2D poses. First, we introduce a cross-view fusion scheme into CNN to jointly estimate 2D poses for multiple views. Consequently, the 2D pose estimation for each view already benefits from other views. Second, we present a recursive Pictorial Structure Model to recover the 3D pose from the multi-view 2D poses. It gradually …
There exist challenging problems in 3D human pose estimation mission, such as poor performance caused by occlusion and self-occlusion. Recently, IMU-vision sensor fusion is regarded as valuable for solving these problems. However, previous researches on the fusion of IMU and vision data, which is heterogeneous, fail to adequately utilize either IMU raw data or reliable high-level vision features. To facilitate a more efficient sensor fusion, in this work we propose a framework called \emph{FusePose} under a parametric human kinematic model. Specifically, we aggregate different information of I…
Southeast University（东南大学）；Microsoft Research Asia（微软亚洲研究院）；The Johns Hopkins University（约翰斯·霍普金斯大学）
Occlusion is probably the biggest challenge for human pose estimation in the wild. Typical solutions often rely on intrusive sensors such as IMUs to detect occluded joints. To make the task truly unconstrained, we present AdaFuse, an adaptive multiview fusion method, which can enhance the features in occluded views by leveraging those in visible views. The core of AdaFuse is to determine the point-point correspondence between two views which we solve effectively by exploring the sparsity of the heatmap representation. We also learn an adaptive fusion weight for each camera view to reflect its …
In this paper, we propose an efficient human pose estimation network -- SFM (slender fusion model) by fusing multi-level features and adding lightweight attention blocks -- HSA (High-Level Spatial Attention). Many existing methods on efficient network have already taken feature fusion into consideration, which largely boosts the performance. However, its performance is far inferior to large network such as ResNet and HRNet due to its limited fusion operation in the network. Specifically, we expand the number of fusion operation by building bridges between two pyramid frameworks without adding …
University of Electronic Science and Technology of China（电子科技大学）
To achieve more accurate 2D human pose estimation, we extend the successful encoder-decoder network, simple baseline network (SBN), in three ways. To reduce the quantization errors caused by the large output stride size, two more decoder modules are appended to the end of the simple baseline network to get full output resolution. Then, the global context blocks (GCBs) are added to the encoder and decoder modules to enhance them with global context features. Furthermore, we propose a novel spatial-attention-based multi-scale feature collection and distribution module (SA-MFCD) to fuse and distr…
Tsinghua University（清华大学）；Carnegie Mellon University（卡内基梅隆大学）；Massachusetts Institute of Technology（麻省理工学院）
Pose estimation plays a critical role in human-centered vision applications. However, it is difficult to deploy state-of-the-art HRNet-based pose estimation models on resource-constrained edge devices due to the high computational cost (more than 150 GMACs per frame). In this paper, we study efficient architecture design for real-time multi-person pose estimation on edge. We reveal that HRNet's high-resolution branches are redundant for models at the low-computation region via our gradual shrinking experiments. Removing them improves both efficiency and performance. Inspired by this finding, w…
A Transformer-based Multimodal Fusion Model for Efficient Crowd Counting Using Visual and Wireless Signals
School of Electrical and Electronic Engineering, University College Dublin（都柏林大学电子与电气工程学院）；College of Electrical Engineering and Automation, Shandong University of Science and Technology（山东科技大学电气工程与自动化学院）
Hangzhou Research Institute, Xidian University（西安电子科技大学杭州研究院）；Xidian University（西安电子科技大学）
For the current 3D human pose estimation task, a group of methods mainly learn the rules of 2D-3D projection from spatial and temporal correlation. However, earlier methods model the global features of the entire body joint in the time domain, but ignore the motion trajectory of individual joint. The recent work [29] considers that there are differences in motion between different joints and deals with the temporal relationship of each joint separately. However, we found that different joints show the same movement trends under some specific actions. Therefore, our proposed Fusionformer method…
Mamba-Driven Topology Fusion for Monocular 3D Human Pose Estimation
College of Sciences, Northeastern University（东北大学理学院）；Key Laboratory of Differential Equations and Their Applications, Northeastern University（东北大学偏微分方程及其应用重点实验室）；Nanjing University of Science and Technology（南京理工大学）
Baidu Research, Baidu Inc.（百度研究部，百度公司）；Hong Kong University of Science and Technology（香港科技大学）；YouTu Lab, Tencent（优图实验室，腾讯）；The Chinese University of Hong Kong（香港中文大学）；University of Kentucky（肯塔基大学）
We explore the importance of spatial contextual information in human pose estimation. Most state-of-the-art pose networks are trained in a multi-stage manner and produce several auxiliary predictions for deep supervision. With this principle, we present two conceptually simple and yet computational efficient modules, namely Cascade Prediction Fusion (CPF) and Pose Graph Neural Network (PGNN), to exploit underlying contextual information. Cascade prediction fusion accumulates prediction maps from previous stages to extract informative signals. The resulting maps also function as a prior to guid…
From Sparse to Dense: Spatio-Temporal Fusion for Multi-View 3D Human Pose Estimation with DenseWarper
Department of Computer Science, THUAI, BNRist, Tsinghua University, Beijing, China（清华大学北京研究院，清华大学计算机科学系，北京，中国）；Dalian University of Technology, Dalian, China（大连理工大学，大连，中国）；Apple, Beijing, China（苹果公司，北京，中国）；Hong Kong University of Science and Technology (Guang Zhou), Guang Zhou, China（香港科技大学（广州），广州，中国）；University of Manchester, Manchester, UK（曼彻斯特大学，曼彻斯特，英国）
University of California, Irvine（加州大学欧文分校）；Tencent Inc（腾讯公司）；Meta AI
Recently, the vision transformer and its variants have played an increasingly important role in both monocular and multi-view human pose estimation. Considering image patches as tokens, transformers can model the global dependencies within the entire image or across images from other views. However, global attention is computationally expensive. As a consequence, it is difficult to scale up these transformer-based methods to high-resolution features and many views. In this paper, we propose the token-Pruned Pose Transformer (PPT) for 2D human pose estimation, which can locate a rough human mas…
In this work, we propose a novel framework named Region-Aware Network (RANet), which learns the ability of anti-confusing in case of heavy occlusion, nearby person and symmetric appearance, for human pose estimation. Specifically, the proposed method addresses three key aspects, i.e., data augmentation, feature learning and prediction fusion, respectively. First, we propose Parsing-based Data Augmentation (PDA) to generate abundant data that synthesizes confusing textures. Second, we not only propose a Feature Pyramid Stem (FPS) to learn stronger low-level features in lower stage; but also inc…
University of Oxford（牛津大学）；University of Leeds（利兹大学）
The objective of this work is human pose estimation in videos, where multiple frames are available. We investigate a ConvNet architecture that is able to benefit from temporal context by combining information across the multiple frames using optical flow. To this end we propose a network architecture with the following novelties: (i) a deeper network than previously investigated for regressing heatmaps; (ii) spatial fusion layers that learn an implicit spatial model; (iii) optical flow is used to align heatmap predictions from neighbouring frames; and (iv) a final parametric pooling layer whic…
Case Western Reserve University（凯斯西储大学）；PAII Inc.（平安智能信息有限公司）；Ping An Technology（平安科技）；Duke Kunshan University（昆山杜克大学）
Transformer-based approaches have been successfully proposed for 3D human pose estimation (HPE) from 2D pose sequence and achieved state-of-the-art (SOTA) performance. However, current SOTAs have difficulties in modeling spatial-temporal correlations of joints at different levels simultaneously. This is due to the poses' spatial-temporal complexity. Poses move at various speeds temporarily with various joints and body-parts movement spatially. Hence, a cookie-cutter transformer is non-adaptable and can hardly meet the "in-the-wild" requirement. To mitigate this issue, we propose Hierarchical S…
OPPO Research Institute（OPPO研究院）；Shanghai JiaoTong University（上海交通大学）；Wuhan University（武汉大学）
Visual localization is a fundamental task that regresses the 6 Degree Of Freedom (6DoF) poses with image features in order to serve the high precision localization requests in many robotics applications. Degenerate conditions like motion blur, illumination changes and environment variations place great challenges in this task. Fusion with additional information, such as sequential information and Inertial Measurement Unit (IMU) inputs, would greatly assist such problems. In this paper, we present an efficient client-server visual localization architecture that fuses global and local pose estim…
Department of Computing, Imperial College London（伦敦帝国理工学院计算机系）
We propose a new multi-instance dynamic RGB-D SLAM system using an object-level octree-based volumetric representation. It can provide robust camera tracking in dynamic environments and at the same time, continuously estimate geometric, semantic, and motion properties for arbitrary objects in the scene. For each incoming frame, we perform instance segmentation to detect objects and refine mask boundaries using geometric and motion information. Meanwhile, we estimate the pose of each existing moving object using an object-oriented tracking method and robustly track the camera pose against the s…
An accurate and robust large-scale localization system is an integral component for active areas of research such as autonomous vehicles and augmented reality. To this end, many learning algorithms have been proposed that predict 6DOF camera pose from RGB or RGB-D images. However, previous methods that incorporate depth typically treat the data the same way as RGB images, often adding depth maps as additional channels to RGB images and passing them through convolutional neural networks (CNNs). In this paper, we show that converting depth maps into pseudo-LiDAR signals, previously shown to be u…
PyCAT4: A Hierarchical Vision Transformer-based Framework for 3D Human Pose Estimation
Department of Computer Science（计算机科学系）；University College London（伦敦大学学院）；School of Electronic Engineering（电子工程学院）；Queen Mary University of London（伦敦女王学院）；Department of Earth Science（地球科学系）；Imperial College London（帝国理工学院）
We propose POse-guided SElective Fusion (POSEFusion), a single-view human volumetric capture method that leverages tracking-based methods and tracking-free inference to achieve high-fidelity and dynamic 3D reconstruction. By contributing a novel reconstruction framework which contains pose-guided keyframe selection and robust implicit surface fusion, our method fully utilizes the advantages of both tracking-based methods and tracking-free inference methods, and finally enables the high-fidelity reconstruction of dynamic surface details even in the invisible regions. We formulate the keyframe s…
DSFFNet: Dual-Side Feature Fusion Network for 3D Pose Transfer
School of Automation, Southeast University（东南大学自动化学院）；Key Laboratory of Measurement and Control of Complex Systems of Engineering, Ministry of Education（复杂工程测量与控制教育部重点实验室）
Benchmarking Visual-Inertial Deep Multimodal Fusion for Relative Pose Regression and Odometry-aided Absolute Pose Regression
Fraunhofer Institute for Integrated Circuits IIS（弗劳恩霍夫集成电路研究所）；Munich Center for Machine Learning（慕尼黑机器学习中心）；Technical University of Dortmund（多特蒙德工业大学）；Ulm University（乌尔姆大学）
Depth map fusion is an essential part in both stereo and RGB-D based 3-D reconstruction pipelines. Whether produced with a passive stereo reconstruction or using an active depth sensor, such as Microsoft Kinect, the depth maps have noise and may have poor initial registration. In this paper, we introduce a method which is capable of handling outliers, and especially, even significant registration errors. The proposed method first fuses a sequence of depth maps into a single non-redundant point cloud so that the redundant points are merged together by giving more weight to more certain measurem…
Tracking sports players is a widely challenging scenario, specially in single-feed videos recorded in tight courts, where cluttering and occlusions cannot be avoided. This paper presents an analysis of several geometric and semantic visual features to detect and track basketball players. An ablation study is carried out and then used to remark that a robust tracker can be built with Deep Learning features, without the need of extracting contextual ones, such as proximity or color similarity, nor applying camera stabilization techniques. The presented tracker consists of: (1) a detection step, …
Panasonic R&D Center（松下研发中心）；Panasonic Corporation（松下公司）；Core Element Technology Development Center（核心元件技术开发中心）
Person re-identification (re-ID) aims to accurately re- trieve a person from a large-scale database of images cap- tured across multiple cameras. Existing works learn deep representations using a large training subset of unique per- sons. However, identifying unseen persons is critical for a good re-ID algorithm. Moreover, the misalignment be- tween person crops to detection errors or pose variations leads to poor feature matching. In this work, we present a fusion of handcrafted features and deep feature representa- tion learned using multiple body parts to complement the global body features…
We propose a novel idea for depth estimation from multi-view image-pose pairs, where the model has capability to leverage information from previous latent-space encodings of the scene. This model uses pairs of images and poses, which are passed through an encoder--decoder model for disparity estimation. The novelty lies in soft-constraining the bottleneck layer by a nonparametric Gaussian process prior. We propose a pose-kernel structure that encourages similar poses to have resembling latent spaces. The flexibility of the Gaussian process (GP) prior provides adapting memory for fusing informa…
Event-based Motion & Appearance Fusion for 6D Object Pose Tracking
Event-driven Perception for Robotics, Istituto Italiano di Tecnologia, Italy（事件驱动感知机器人实验室，意大利理工学院）；Humanoid Sensing and Perception, Istituto Italiano di Tecnologia, Italy（人形感知与感知，意大利理工学院）；University of Genoa, Genoa, Italy（热那亚大学，意大利）
Institute of Software, Chinese Academy of Sciences（中国科学院软件研究所）；University of Chinese Academy of Sciences（中国科学院大学）；Alibaba（阿里巴巴）；Simon Fraser University（西蒙弗雷泽大学）；Google（谷歌）
3D hand pose estimation from single depth is a fundamental problem in computer vision, and has wide applications.However, the existing methods still can not achieve satisfactory hand pose estimation results due to view variation and occlusion of human hand. In this paper, we propose a new virtual view selection and fusion module for 3D hand pose estimation from single depth.We propose to automatically select multiple virtual viewpoints for pose estimation and fuse the results of all and find this empirically delivers accurate and robust pose estimation. In order to select most effective virtua…
Cross-Modal RGB-D Fusion Transformer for 6D Pose Estimation of Non-Cooperative Spacecraft with Stereo-Derived Depth
School of Physics, Northeast Normal University（东北师范大学物理学院）；Changchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Science（长春光学精密机械与物理研究所）
