Research map / 3D vision / 3D generation
Learning Environment-Aware Affordance for 3D Articulated Object Manipulation under Occlusions
Ruihai Wu, Kai Cheng, Yan Zhao, Chuanruo Ning, Guanqi Zhan, Hao Dong
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 |
|---|---|---|---|
| remove occlusion field (ablation) | Ours · Pushing | sample manipulation accuracy: 39.43 -> 43.52 | Improved |
| remove occlusion field (ablation) | Ours · Pulling | sample manipulation accuracy: 33.19 -> 36.19 | Improved |
| remove contrastive learning (ablation) | Ours · Pushing | sample manipulation accuracy: 36.63 -> 43.52 | Improved |
| remove contrastive learning (ablation) | Ours · Pulling | sample manipulation accuracy: 28.02 -> 36.19 | Improved |
| remove occlusion field (ablation) | Ours · Pulling F-score | F-score: 36.36 -> 71.28 | Improved |
| remove occlusion field (ablation) | Ours · Pushing F-score | F-score: 64.28 -> 77.30 | Improved |
| remove contrastive learning (ablation) | Ours · Pulling (novel) F-score | F-score: 55.25 -> 61.32 | Improved |
| remove contrastive learning (ablation) | Ours · Pushing F-score | F-score: 68.59 -> 77.30 | Improved |
| remove contrastive learning (ablation) | Ours · Pulling average precision | average precision: 75.40 -> 75.96 | No clear effect |
| train O2O-M on multi-occluder scenes vs Ours trained data-efficiently | O2O-M · Pushing | sample manipulation accuracy: 37.33 -> 43.52 | Improved |
| replace flying gripper (W2A) with robot arm (W2A-R) | Where2Act · Pushing | sample manipulation accuracy: 17.80 -> 34.76 | Improved |
| replace flying gripper (W2A) with robot arm (W2A-R) | Where2Act · Pulling | sample manipulation accuracy: 3.82 -> 27.55 | Improved |
Closest papers by meaning
- IAAO: Interactive Affordance Learning for Articulated Objects in 3D Environments (CVPR 2025 · similarity 0.92)
- AffordDP: Generalizable Diffusion Policy with Transferable Affordance (CVPR 2025 · similarity 0.90)
- Grounding 3D Object Affordance with Language Instructions, Visual Observations and Interactions (CVPR 2025 · similarity 0.89)
- GREAT: Geometry-Intention Collaborative Inference for Open-Vocabulary 3D Object Affordance Grounding (CVPR 2025 · similarity 0.88)
- Learning 2D Invariant Affordance Knowledge for 3D Affordance Grounding (AAAI 2025 · similarity 0.88)
- VidBot: Learning Generalizable 3D Actions from In-the-Wild 2D Human Videos for Zero-Shot Robotic Manipulation (CVPR 2025 · similarity 0.88)
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.
