Research map / 3D vision / Human pose estimation
Explicit Occlusion Reasoning for Multi-Person 3D Human Pose Estimation
Qihao Liu, Yi Zhang, Song Bai, Alan Yuille
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add occluded-keypoint reasoning module to detection (ablation) | HUPOR · MuPoTS-3D | PCK_rel: 74.87 -> 79.28 | Improved |
| add refinement module after reasoning (ablation) | HUPOR · MuPoTS-3D | PCK_rel: 79.28 -> 86.54 | Improved |
| train detection on visible joints only (use occlusion labels) (ablation) | HUPOR detection · MuPoTS-3D | PCK_rel: 71.12 -> 74.87 | Improved |
| train detection on visible joints only (use occlusion labels) (ablation) | HUPOR detection · MuPoTS-3D | PCK_occ: 56.53 -> 51.48 | Hurt |
| use occlusion labels when reasoning module is added (ablation) | HUPOR · MuPoTS-3D | PCK_rel: 74.76 -> 79.28 | Improved |
| replace hourglass reasoning with DSED (ablation) | HUPOR · MuPoTS-3D | PCK_rel: 75.58 -> 79.28 | Improved |
| replace hourglass reasoning with DSED (ablation) | HUPOR · MuPoTS-3D | PCK_occ: 55.30 -> 61.24 | Improved |
| add non-local blocks to DSED (ablation) | HUPOR · MuPoTS-3D | PCK_rel: 79.28 -> 79.81 | No clear effect |
| add DSED reasoning module (ablation) | HrHRNet-W48 · COCO val-dev 2017 | AP: 69.9 -> 70.8 | Improved |
| add hourglass reasoning module (ablation) | HrHRNet-W48 · COCO val-dev 2017 | AP: 69.9 -> 68.2 | Hurt |
| add hourglass reasoning module (ablation) | PifPaf · COCO val-dev 2017 | AP: 67.4 -> 67.5 | No clear effect |
| add DSED reasoning module (ablation) | PifPaf · COCO val-dev 2017 | AP: 67.4 -> 69.1 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- PifPaf — PifPaf: Composite Fields for Human Pose Estimation (CVPR 2019) · better on 1 of 1 shared result
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
- Occlusion-Aware Networks for 3D Human Pose Estimation in Video (ICCV 2019 · similarity 0.91)
- 3D Human Pose Estimation Using Spatio-Temporal Networks with Explicit Occlusion Training (AAAI 2020 · similarity 0.91)
- Human Pose Recognition via Occlusion-Preserving Abstract Images (ECCV 2024 · similarity 0.90)
- Object-Occluded Human Shape and Pose Estimation With Probabilistic Latent Consistency (TPAMI 2022 · similarity 0.90)
- VOccl3D: A Video Benchmark Dataset for 3D Human Pose and Shape Estimation under real Occlusions (ICCV 2025 · similarity 0.89)
- $\text{Di}^2\text{Pose}$: Discrete Diffusion Model for Occluded 3D Human Pose Estimation (NeurIPS 2024 · similarity 0.89)
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.
