Research map / 3D vision / Human pose estimation
Learning Structure-Supporting Dependencies via Keypoint Interactive Transformer for General Mammal Pose Estimation
Tianyang Xu, Jiyong Rao, Xiaoning Song, Zhenhua Feng, Xiaojun Wu
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 adaptive weighting strategy (ablation) | HRNet-W32 · AP10K val | AP: 73.8 -> 74.5 | Improved |
| add KIT module (ablation) | HRNet-W32 · AP10K val | AP: 73.8 -> 74.3 | Improved |
| combine adaptive weighting and KIT (ablation) | HRNet-W32 · AP10K val | AP: 73.8 -> 76.1 | Improved |
| add body part prompt to KIT (ablation) | KITPose · AP10K val | AP: 73.8 -> 75.8 | Improved |
| add body part prompt to adaptive weighting plus KIT (ablation) | KITPose · AP10K val | AP: 76.1 -> 76.6 | Improved |
| Euclidean instead of cosine distance for prompt clustering (ablation) | KITPose-E2 · AP10K val | AP: 75.8 -> 76.6 | Improved |
| tight clustering threshold 1e-8 instead of 1e-2 (ablation) | KITPose-E2 · AP10K val | AP: 75.0 -> 76.6 | Improved |
| number of body part prompts from 3 to 4 (ablation) | KITPose-E2 · AP10K val | AP: 75.8 -> 76.6 | Mixed |
| constrained keypoint weighting instead of hand-crafted (ablation) | HRNet-W32 · AP10K val | AP: 73.8 -> 74.6 | Improved |
| adaptive instead of hand-crafted keypoint weighting (ablation) | KITPose-E2 · AP10K val | AP: 75.0 -> 76.1 | Improved |
| 5x5 Laplacian kernel instead of 3x3 (ablation) | KITPose-E2 · AP10K val | AP: 76.6 -> 76.8 | Improved |
| 7x7 Laplacian kernel instead of 3x3 (ablation) | KITPose-E2 · AP10K val | AP: 76.6 -> 71.5 | Hurt |
Closest papers by meaning
- AP-10K: A Benchmark for Animal Pose Estimation in the Wild (NeurIPS 2021 · similarity 0.90)
- Animal3D: A Comprehensive Dataset of 3D Animal Pose and Shape (ICCV 2023 · similarity 0.89)
- Cross-Domain Adaptation for Animal Pose Estimation (ICCV 2019 · similarity 0.89)
- Pose Splatter: A 3D Gaussian Splatting Model for Quantifying Animal Pose and Appearance (NeurIPS 2025 · similarity 0.88)
- Probabilistic Prompt Distribution Learning for Animal Pose Estimation (CVPR 2025 · similarity 0.87)
- Reconstructing Animals and the Wild (CVPR 2025 · 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.
