Research map / 3D vision / Face and shape
Coarse-to-fine Animal Pose and Shape Estimation
Chen Li, Gim Hee Lee
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 local features to MRGCN input (ablation) | MRGCN · StanfordExtra | IOU: 73.3 -> 81.6 | Improved |
| encoder-decoder MRGCN instead of traditional GCN (ablation) | MRGCN · StanfordExtra | IOU: 79.4 -> 81.6 | Improved |
| Tversky loss instead of L1 loss (ablation) | Full pipeline · StanfordExtra | IOU: 81.1 -> 81.6 | Improved |
| Tversky loss instead of L1 loss (ablation) | Full pipeline · StanfordExtra | PCK@0.15 Tail: 65.2 -> 63.7 | Hurt |
| coarse-to-fine refinement vs WLDO regression baseline | Ours · BADJA | IOU: 65.0 -> 72.0 | Improved |
| coarse-to-fine refinement vs WLDO | Ours · Animal Pose | IOU: 67.5 -> 75.7 | Improved |
| refinement vs coarse estimation under domain shift | Ours · BADJA | PCK@0.15 Avg: 42.5 -> 54.1 | Improved |
| regression-based refinement vs test-time optimization with 200 iterations | Ours · StanfordExtra | IOU: 79.0 -> 81.6 | Improved |
| more test-time optimization iterations | Coarse + optimization · StanfordExtra | Avg PCK: 78.3 -> 81.7 | Improved |
| Gaussian noise on SMAL parameters, 10% to 50% of sigma (ablation) | Ours · StanfordExtra | Avg PCK: 81.3 -> 78.1 | Mixed |
| Gaussian noise on camera parameters, 10% to 50% of sigma (ablation) | Ours · StanfordExtra | Avg PCK: 82.2 -> 69.7 | Hurt |
Closest papers by meaning
- Animal Avatars: Reconstructing Animatable 3D Animals from Casual Videos (ECCV 2024 · similarity 0.89)
- BARC: Breed-Augmented Regression Using Classification for 3D Dog Reconstruction from Images (IJCV 2023 · similarity 0.88)
- Animal3D: A Comprehensive Dataset of 3D Animal Pose and Shape (ICCV 2023 · similarity 0.88)
- VAREN: Very Accurate and Realistic Equine Network (CVPR 2024 · similarity 0.87)
- Pose Splatter: A 3D Gaussian Splatting Model for Quantifying Animal Pose and Appearance (NeurIPS 2025 · similarity 0.86)
- SAM 3D Animal: Promptable Animal 3D Reconstruction from Images in the Wild (NeurIPS 2026 · 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.
