Research map / Image generation / Text-to-image diffusion
Harnessing Text-to-Image Diffusion Models for Category-Agnostic Pose Estimation
Duo Peng, Zhengbo Zhang, Ping Hu, Qiuhong Ke, David Yau, Jun Liu
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 category-shared prompt training (CPT) to PPM (ablation) | PPM · MP-100 Cross Super-Category (1-shot), Mean | PCK@0.2: 68.68 -> 70.06 | Improved |
| add CPT to PPM (ablation) | PPM · MP-100 Cross Sub-Category (5-shot), Mean | PCK@0.2: 90.17 -> 92.58 | Improved |
| average cross-attention maps over multiple U-Net layers instead of best single layer (ablation) | PPM + CPT · MP-100 Cross Super-Category (1-shot) | PCK@0.2: 68.81 -> 70.06 | Improved |
| use learned pseudo prompts on frozen Stable Diffusion with no base-category training | PPM · MP-100 Cross Super-Category (1-shot), Furniture | PCK@0.2: 52.49 -> 58.61 | Improved |
| use PPM instead of CapeFormer on cross sub-category 1-shot | PPM · MP-100 Cross Sub-Category (1-shot), Mean | PCK@0.2: 85.31 -> 85.82 | No clear effect |
Closest papers by meaning
- Unleashing the Power of Text-to-Image Diffusion Models for Category-Agnostic Pose Estimation (TPAMI 2026 · similarity 0.96)
- ESCAPE: Encoding Super-keypoints for Category-Agnostic Pose Estimation (CVPR 2024 · similarity 0.91)
- CapeX: Category-Agnostic Pose Estimation from Textual Point Explanation (ICLR 2025 · similarity 0.91)
- Meta-Point Learning and Refining for Category-Agnostic Pose Estimation (CVPR 2024 · similarity 0.90)
- Object Pose Estimation via the Aggregation of Diffusion Features (CVPR 2024 · similarity 0.86)
- AP-Adapter: Improving Generalization of Automatic Prompts on Unseen Text-to-Image Diffusion Models (NeurIPS 2024 · similarity 0.86)
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.
