Research map / 3D vision / 3D generation
2HandedAfforder: Learning Precise Actionable Bimanual Affordances from Human Videos
Marvin Heidinger, Snehal Jauhri, Vignesh Prasad, Georgia Chalvatzaki
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 |
|---|---|---|---|
| use VLM [SEG] token instead of CLIP text features (ablation) | 2HandedAfforder-CLIP · ActAffordance (combined) | Precision: 0.064 -> 0.130 | Improved |
| use VLM [SEG] token instead of CLIP text features (ablation) | 2HandedAfforder-CLIP · ActAffordance-Cropped (combined) | mAP: 0.154 -> 0.240 | Improved |
| train on 2HANDS with same LISA-like architecture | LISA · ActAffordance (combined) | Precision: 0.050 -> 0.130 | Improved |
| train on 2HANDS with same architecture as AffordanceLLM | AffordanceLLM · ActAffordance (combined) | IoU: 0.012 -> 0.058 | Improved |
| crop benchmark images to target-object boxes | 2HandedAfforder · ActAffordance | Precision: 0.130 -> 0.269 | Improved |
| crop benchmark images to target-object boxes | LISA · ActAffordance | Precision: 0.050 -> 0.122 | Improved |
| evaluate on unseen Ego4D split rather than EPIC-KITCHENS | 2HandedAfforder · ActAffordance | Precision: 0.125 -> 0.137 | No clear effect |
| use SAM2 video segmentation mask completion instead of MI-GAN image reconstruction (ablation) | affordance extraction · 2HANDS extraction | qualitative mask accuracy | Improved |
| use both-hand affordance regions for robot stirring instead of single arm or plain object segmentation | Tiago++ robot · stirring vegetables | task success | Improved |
| use AffExtract extraction directly on benchmark sequences | AffExtract · ActAffordance (combined) | Precision: 0.130 -> 0.420 | Improved |
Closest papers by meaning
- Learning Precise Affordances from Egocentric Videos for Robotic Manipulation (ICCV 2025 · similarity 0.92)
- Learning 2D Invariant Affordance Knowledge for 3D Affordance Grounding (AAAI 2025 · similarity 0.89)
- OAKINK2: A Dataset of Bimanual Hands-Object Manipulation in Complex Task Completion (CVPR 2024 · similarity 0.89)
- Weakly Supervised Multimodal Affordance Grounding for Egocentric Images (AAAI 2024 · similarity 0.88)
- SeqAfford: Sequential 3D Affordance Reasoning via Multimodal Large Language Model (CVPR 2025 · similarity 0.88)
- IAAO: Interactive Affordance Learning for Articulated Objects in 3D Environments (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.
