Research map / 3D vision / Face and shape
3D Bird Reconstruction: a Dataset, Model, and Shape Recovery from a Single View
Marc Badger, Yufu Wang, Adarsh Modh, Ammon Perkes, Nikos Kolotouros, Bernd G. Pfrommer, Marc F. Schmidt, Kostas Daniilidis
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 mask term (10:1 weight) to keypoint-only objective (ablation) | multi-view optimization · Multi-view fits | IoU: 0.540 -> 0.560 | Improved |
| add mask term (1:1 weight) to keypoint-only objective (ablation) | multi-view optimization · Multi-view fits | IoU: 0.540 -> 0.624 | Improved |
| add mask term (1:1 weight) to keypoint-only objective (ablation) | multi-view optimization · Multi-view fits | PCK@05: 0.356 -> 0.328 | Hurt |
| add mask term (10:1 weight) to keypoint-only objective (ablation) | multi-view optimization · Multi-view fits | PCK@10: 0.631 -> 0.637 | No clear effect |
| initialize single-view optimization with pose regression network (ablation) | single-view optimization · Single-view same-view evaluation | PCK@05: 0.331 -> 0.364 | Improved |
| initialize single-view optimization with pose regression network (ablation) | single-view optimization · Single-view same-view evaluation | IoU: 0.641 -> 0.671 | Improved |
| add single-view optimization after pose regression (ablation) | pose regression network · Single-view same-view evaluation | PCK@05: 0.104 -> 0.364 | Improved |
| use ground truth keypoints and masks as pipeline input instead of predictions | single-view pipeline · Cross-view evaluation | PCK@05: 0.313 -> 0.332 | Improved |
| use ground truth keypoints and masks as pipeline input instead of predictions | single-view pipeline · Cross-view evaluation | IoU: 0.589 -> 0.586 | No clear effect |
| remove pose prior and bone limit terms from single-view objective (ablation) | single-view optimization · Single-view fitting | PCK: PCK increased; IoU decreased | Mixed |
Closest papers by meaning
- Birds of a Feather: Capturing Avian Shape Models From Images (CVPR 2021 · similarity 0.91)
- Multi-view Tracking, Re-ID, and Social Network Analysis of a Flock of Visually Similar Birds in an Outdoor Aviary (IJCV 2023 · similarity 0.90)
- Pose Splatter: A 3D Gaussian Splatting Model for Quantifying Animal Pose and Appearance (NeurIPS 2025 · similarity 0.88)
- SAM 3D Animal: Promptable Animal 3D Reconstruction from Images in the Wild (NeurIPS 2026 · similarity 0.87)
- BEAST3D: animal behavioral analysis and neural encoding from multi-view video via Gaussian splatting (NeurIPS 2026 · similarity 0.87)
- Web-Scale Collection of Video Data for 4D Animal Reconstruction (NeurIPS 2025 · similarity 0.87)
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.
