Research map / Label-efficient and robust learning / Self-supervised and contrastive
Beyond Appearance: A Semantic Controllable Self-Supervised Learning Framework for Human-Centric Visual Tasks
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 semantic clustering supervision (ablation) | DINO Swin-T · MSMT17 | mAP/Rank1: 63.3/83.2 -> 61.6/82.2 | Hurt |
| Add semantic clustering supervision (ablation) | DINO Swin-T · Market1501 | mAP/Rank1: 89.6/95.9 -> 89.5/95.5 | Hurt |
| Add semantic clustering supervision (ablation) | DINO Swin-T · PA100k | mA: 82.98 -> 82.98 | No clear effect |
| Add semantic clustering supervision (ablation) | DINO Swin-T · LIP | mIoU: 54.45 -> 55.25 | Helped |
| Add semantic clustering supervision (ablation) | DINO Swin-T · CityPerson | MR^-2(R/HO): 11.4/43.1 -> 11.1/41.7 | Helped |
| Add semantic clustering supervision (ablation) | DINO Swin-T · PRW | mAP/Rank1: 52.9/84.7 -> 53.0/84.0 | Mixed |
| Add semantic controller to semantic clustering (ablation) | SOLIDER · Market1501 | mAP/Rank1: 89.5/95.5 -> 89.9/96.1 | Helped |
| Add semantic controller to semantic clustering (ablation) | SOLIDER · MSMT17 | mAP/Rank1: 61.6/82.2 -> 63.9/83.8 | Helped |
| Add semantic controller to semantic clustering (ablation) | SOLIDER · PRW | mAP/Rank1: 53.0/84.0 -> 54.1/85.0 | Mixed |
| Sample the semantic controller value from a binary distribution instead of a uniform or beta distribution (ablation) | SOLIDER · downstream human-centric tasks | performance | Helped |
| Increase the semantic controller value (ablation) | SOLIDER · person re-identification and pedestrian detection | performance | Mixed |
Closest papers by meaning
- Can Semantic Labels Assist Self-Supervised Visual Representation Learning? (AAAI 2022 · similarity 0.86)
- Modeling Collaborator: Enabling Subjective Vision Classification With Minimal Human Effort via LLM Tool-Use (CVPR 2024 · similarity 0.83)
- Probing the Mid-level Vision Capabilities of Self-Supervised Learning (CVPR 2025 · similarity 0.83)
- L-Man: A Large Multi-modal Model Unifying Human-centric Tasks (AAAI 2025 · similarity 0.83)
- Cross-view and Cross-pose Completion for 3D Human Understanding (CVPR 2024 · similarity 0.83)
- PLIP: Language-Image Pre-training for Person Representation Learning (NeurIPS 2024 · similarity 0.83)
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.