AI Paper Atlas

Research map / 3D vision / Human pose estimation

VoxelTrack: Multi-Person 3D Human Pose Estimation and Tracking in the Wild

Yifu Zhang, Chunyu Wang, Xinggang Wang, Wenyu Liu, Wenjun Zeng

TPAMI 2022 · 3D vision › Human pose estimation · approach: Multi-person pose estimation · 99 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Increase voxel size from 80x80x32 to 120x120x48 (ablation)VoxelTrack · PanopticAP25: 39.74 -> 71.00Improved
Increase voxel size from 120x120x48 to 160x160x64 (ablation)VoxelTrack · PanopticAP25: 71.00 -> 79.34Improved
Increase voxel size from 120x120x48 to 160x160x64 (ablation)VoxelTrack · PanopticMPJPE: 19.83 mm -> 18.49 mmImproved
Use sparse convolution instead of standard convolution (160x160x64) (ablation)VoxelTrack · PanopticJEN time: 132.64 ms -> 48.46 msImproved
Use sparse convolution instead of standard convolution (80x80x32) (ablation)VoxelTrack · PanopticJEN time: 18.75 ms -> 22.01 msHurt
Use 5 cameras instead of 3 (ablation)VoxelTrack · PanopticAP25: 49.09 -> 79.34Improved
Use 5 cameras instead of 3 (ablation)VoxelTrack · PanopticID Switch: 0 -> 0No clear effect
Use Higher-HRNet-W32 backbone instead of DLA-34 (ablation)VoxelTrack · PanopticAP25: 79.34 -> 85.88Improved
Use MobileNet-V2 backbone instead of DLA-34 (ablation)VoxelTrack · PanopticAP25: 79.34 -> 42.42Hurt
Use larger input image 960x512 instead of 640x384 (ablation)VoxelTrack · PanopticAP25: 55.96 -> 79.34Improved
Add Re-ID features to 3D-pose-only tracking (ablation)VoxelTrack · PanopticID Switch: 90 -> 15Improved
Add occlusion mask when fusing Re-ID features (ablation)VoxelTrack · PanopticIDF1: 94.38 -> 98.67Improved

Later papers that compared with this method

1 later accepted paper reports VoxelTrack as a baseline in their results tables; 0 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy