Research map / Detection and segmentation / Weakly supervised detection and segmentation
Selective Contrastive Learning for Weakly Supervised Affordance Grounding
WonJun Moon, Hyun Seok Seong, Jae-Pil Heo
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 object-level prototypical contrastive learning (ablation) | classification baseline · AGD20K-Seen | KLD: 1.349 -> 1.271 | Improved |
| add object-level pixel contrastive learning (ablation) | object-level proto model · AGD20K-Seen | KLD: 1.271 -> 1.219 | Improved |
| add part-level prototypical contrastive learning (ablation) | object-level proto + pixel model · AGD20K-Seen | KLD: 1.219 -> 1.164 | Improved |
| add part-level pixel contrastive learning (ablation) | object+part proto, object pixel · AGD20K-Seen | KLD: 1.157 -> 1.142 | Improved |
| add CAM calibration with object affinity map (ablation) | full contrastive model · AGD20K-Seen | KLD: 1.142 -> 1.124 | Improved |
| add CAM calibration with object affinity map (ablation) | classification baseline · AGD20K-Seen | KLD: 1.349 -> 1.271 | Improved |
| add object-level pixel contrastive learning on top of object+part proto (ablation) | object+part proto model · AGD20K-Seen | SIM: 0.416 -> 0.414 | No clear effect |
| replace CLIP with ClearCLIP (ablation) | DINO+CLIP model · AGD20K-Seen | KLD: 1.160 -> 1.124 | Improved |
| use prompt 'an item to {action} with' instead of bare action (ablation) | zero-shot ClearCLIP · AGD20K-Seen | KLD: 1.672 -> 1.574 | Improved |
| use CLIP affinity map instead of DINO attention for pIoU (ablation) | full model · AGD20K-Unseen | KLD: 1.243 -> 1.257 | Hurt |
| use CLIP affinity map instead of DINO attention for pIoU (ablation) | full model · AGD20K-Seen | KLD: 1.124 -> 1.126 | No clear effect |
| selective contrastive learning vs LOCATE | LOCATE · AGD20K-Unseen | KLD: 1.405 -> 1.243 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- LOCATE — LOCATE: Localize and Transfer Object Parts for Weakly Supervised Affordance Grounding (CVPR 2023) · better on 6 of 6 shared results
- INTRA — INTRA: Interaction Relationship-aware Weakly Supervised Affordance Grounding (ECCV 2024) · better on 2 of 2 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
- Weakly-Supervised Affordance Grounding Guided by Part-Level Semantic Priors (ICLR 2025 · similarity 0.93)
- Learning Affordance Grounding From Exocentric Images (CVPR 2022 · similarity 0.93)
- Weakly Supervised Multimodal Affordance Grounding for Egocentric Images (AAAI 2024 · similarity 0.92)
- Learning 2D Invariant Affordance Knowledge for 3D Affordance Grounding (AAAI 2025 · similarity 0.89)
- Reasoning Mamba: Hypergraph-Guided Region Relation Calculating for Weakly Supervised Affordance Grounding (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.
