Research map / 3D vision / Human pose estimation
Detailed 2D-3D Joint Representation for Human-Object Interaction
Yong-Lu Li, Xinpeng Liu, Han Lu, Shiyi Wang, Junqi Liu, Jiefeng Li, Cewu Lu
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 |
|---|---|---|---|
| use DJ-RN's detailed 2D-3D joint representation | HOI detection · HICO-DET Default Full | mAP: 19.40 -> 21.34 | Improved |
| use DJ-RN's detailed 2D-3D joint representation | HOI detection · Ambiguous-HOI | mAP: 9.72 -> 10.37 | Improved |
| replace VPoser embedding with 3D pose in the 3D body block (ablation) | DJ-RN · HICO-DET Default Full | mAP: 21.34 -> 20.42 | Hurt |
| replace VPoser embedding with point cloud in the 3D body block (ablation) | DJ-RN · HICO-DET Default Full | mAP: 21.34 -> 20.05 | Hurt |
| add detailed face shape (ablation) | DJ-RN · HICO-DET Default Full | mAP: 21.02 -> 21.34 | Improved |
| add detailed hand shape (ablation) | DJ-RN · HICO-DET Default Full | mAP: 20.83 -> 21.34 | Improved |
| add both detailed face and hand shapes (ablation) | DJ-RN · HICO-DET Default Full | mAP: 20.74 -> 21.34 | Improved |
| add the 3D volume block (ablation) | DJ-RN · HICO-DET Default Full | mAP: 20.34 -> 21.34 | Improved |
| add the 3D body block (ablation) | DJ-RN · HICO-DET Default Full | mAP: 20.01 -> 21.34 | Improved |
| add attention consistency loss (ablation) | DJ-RN · HICO-DET Default Full | mAP: 20.70 -> 21.34 | Improved |
| add spatial triplet loss (ablation) | DJ-RN · HICO-DET Default Full | mAP: 20.83 -> 21.34 | Improved |
| add semantic consistency loss (ablation) | DJ-RN · HICO-DET Default Full | mAP: 20.80 -> 21.34 | Improved |
Closest papers by meaning
- Exploring Pose-Aware Human-Object Interaction via Hybrid Learning (CVPR 2024 · similarity 0.89)
- InterAct: Advancing Large-Scale Versatile 3D Human-Object Interaction Generation (CVPR 2025 · similarity 0.89)
- Reconstructing In-the-Wild Open-Vocabulary Human-Object Interactions (CVPR 2025 · similarity 0.88)
- HOI-M^3: Capture Multiple Humans and Objects Interaction within Contextual Environment (CVPR 2024 · similarity 0.88)
- HORP: Human-Object Relation Priors Guided HOI Detection (CVPR 2025 · similarity 0.87)
- InteractAnything: Zero-shot Human Object Interaction Synthesis via LLM Feedback and Object Affordance Parsing (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.
