Research map / 3D vision / Human pose estimation
Bring Your Rear Cameras for Egocentric 3D Human Pose Estimation
Hiroyasu Akada, Jian Wang, Vladislav Golyanik, Christian Theobalt
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 the proposed refinement module with rear views | EgoPoseFormer · Ego4View-RW (2 front + 2 rear views) | MPJPE: 63.38 -> 56.94 | Improved |
| add the proposed refinement module with rear views | EgoTAP · Ego4View-RW (2 front + 2 rear views) | MPJPE: 69.78 -> 62.11 | Improved |
| add the proposed method with only front views | EgoPoseFormer · Ego4View-Syn (2 front views) | MPJPE: 27.36 -> 27.04 | No clear effect |
| add rear views (2 rear) to 2 front views | EgoPoseFormer · Ego4View-Syn | MPJPE: 27.36 -> 20.20 | Improved |
| add 1 rear-left view to 2 front views | EgoPoseFormer · Ego4View-RW | MPJPE: 77.95 -> 66.08 | Improved |
| add deformable attention guided by 2D anchors instead of cross attention (ablation) | Refinement module · Ego4View-Syn (2 front + 2 rear views) | MPJPE: 20.05 -> 19.25 | Improved |
| add deformable attention guided by 2D anchors instead of cross attention (ablation) | Refinement module · Ego4View-Syn (2 front + 2 rear views) | rear-view heatmap MSE: 2.76 -> 1.83 | Improved |
| add uncertainty-aware masking (ablation) | Refinement module · Ego4View-Syn (2 front + 2 rear views) | MPJPE: 19.67 -> 19.25 | Improved |
| add RGB and heatmap embeddings (ablation) | Refinement module · Ego4View-Syn (2 front + 2 rear views) | MPJPE: 19.47 -> 19.25 | Improved |
| add RGB embedding (ablation) | Refinement module · Ego4View-Syn (2 front + 2 rear views) | MPJPE: 19.45 -> 19.25 | Improved |
| add the refinement module over baseline heatmaps (ablation) | EgoPoseFormer · Ego4View-Syn (2 front + 2 rear views) | rear-view heatmap MSE: 3.36 -> 1.83 | 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.
- EgoPoseFormer — EgoPoseFormer: A Simple Baseline for Stereo Egocentric 3D Human Pose Estimation (ECCV 2024) · better on 8 of 8 shared results
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
- 3D Human Pose Perception from Egocentric Stereo Videos (CVPR 2024 · similarity 0.90)
- EgoSim: An Egocentric Multi-view Simulator and Real Dataset for Body-worn Cameras during Motion and Activity (NeurIPS 2024 · similarity 0.89)
- FRAME: Floor-aligned Representation for Avatar Motion from Egocentric Video (CVPR 2025 · similarity 0.89)
- Instance Tracking in 3D Scenes from Egocentric Videos (CVPR 2024 · similarity 0.88)
- HMD-Poser: On-Device Real-time Human Motion Tracking from Scalable Sparse Observations (CVPR 2024 · similarity 0.88)
- HOT3D: Hand and Object Tracking in 3D from Egocentric Multi-View Videos (CVPR 2025 · similarity 0.87)
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.
