Abstract:Monocular 3D human pose estimation has made progress in recent years. Most of the methods focus on single persons, which estimate the poses in the person-centric coordinates, i.e., the coordinates based on the center of the target person. Hence, these methods are inapplicable for multi-person 3D pose estimation, where the absolute coordinates (e.g., the camera coordinates) are required. Moreover, multi-person pose estimation is more challenging than single pose estimation, due to inter-person occlusion and close human interactions. Existing top-down multi-person methods rely on human detection (i.e., top-down approach), and thus suffer from the detection errors and cannot produce reliable pose estimation in multi-person scenes. Meanwhile, existing bottom-up methods that do not use human detection are not affected by detection errors, but since they process all persons in a scene at once, they are prone to errors, particularly for persons in small scales. To address all these challenges, we propose the integration of top-down and bottom-up approaches to exploit their strengths. Our top-down network estimates human joints from all persons instead of one in an image patch, making it robust to possible erroneous bounding boxes. Our bottom-up network incorporates human-detection based normalized heatmaps, allowing the network to be more robust in handling scale variations. Finally, the estimated 3D poses from the top-down and bottom-up networks are fed into our integration network for final 3D poses. To address the common gaps between training and testing data, we do optimization during the test time, by refining the estimated 3D human poses using high-order temporal constraint, re-projection loss, and bone length regularizations. Our evaluations demonstrate the effectiveness of the proposed method. Code and models are available: <a class="link-external link-https" href="https://github.com/3dpose/3D-Multi-Person-Pose" rel="external noopener nofollow">this https URL</a>.

Multi-Person Pose Estimation with LIMB Detection Heatmaps

Multi-Person Pose Estimation with Accurate Heatmap Regression and Greedy Association

Exploring Severe Occlusion: Multi-Person 3D Pose Estimation with Gated Convolution.

Unsupervised Universal Hierarchical Multi-Person 3D Pose Estimation for Natural Scenes

DeepCut: Joint Subset Partition and Labeling for Multi Person Pose Estimation

Single-view Multi-Human Pose Estimation by Attentive Cross-Dimension Matching.

Top-Down Meets Bottom-Up for Multi-Person Pose Estimation

Multi-Person Pose Estimation Using Bounding Box Constraint and LSTM.

Multi-person Pose Estimation with Object Occlusion Information

Monocular 3D Multi-Person Pose Estimation by Integrating Top-Down and Bottom-Up Networks

Human Pose Estimation from Depth Images via Inference Embedded Multi-task Learning

Human Pose Estimation Based on Lightweight Multi-Scale Coordinate Attention

Dual networks based 3D Multi-Person Pose Estimation from Monocular Video

PoseDet: Fast Multi-Person Pose Estimation Using Pose Embedding

Graph-Based 3D Multi-Person Pose Estimation Using Multi-View Images

Multi-Person 3d Pose Estimation From Monocular Image Sequences

Multi-person Pose Estimation Based on Graph Grouping Optimization

3D Human Pose Estimation from Deep Multi-View 2D Pose

Multi-Person Full Body Pose Estimation

A Deconvolutional Bottom-up Deep Network for Multi-Person Pose Estimation.

Multi-Person Pose Estimation in the Wild: Using Adversarial Method to Train a Top-Down Pose Estimation Network