Research map / 3D vision / Face and shape
Learning Monocular 3D Reconstruction of Articulated Categories From Motion
Check 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 motion-based training supervision (ablation) | Horse reconstruction inference · Horses | mIoU: 0.519 -> 0.536 | Helped |
| Add motion-based training supervision (ablation) | Horse reconstruction inference · Horses | PCK: 71.5 -> 74.7 | Helped |
| Add motion-based terms to mask refinement (ablation) | Horse reconstruction · Horses | PCK: 79.5 -> 83.1 | Helped |
| Add motion-based terms to mask refinement (ablation) | Horse reconstruction · Horses | mIoU: 0.691 -> 0.631 | Hurt |
| Add test-time refinement | Ours, Horse TigDog with keypoints · Horse TigDog, with keypoints | PCK: 74.7 -> 83.1 | Helped |
| Add test-time refinement | Ours, Horse Pascal with keypoints · Horse Pascal, with keypoints | PCK: 57.2 -> 69.5 | Helped |
| Add test-time refinement | Ours, Tiger TigDog with keypoints · Tiger TigDog, with keypoints | PCK: 51.9 -> 55.7 | Helped |
| Add test-time refinement | Ours, mask-only Horse TigDog · Horse TigDog, mask-only training | PCK: 42.5 -> 61.3 | Helped |
| Add test-time refinement | Ours, mask-only Horse Pascal · Horse Pascal, mask-only training | PCK: 31.6 -> 54.9 | Helped |
| Increase the number of deformation handles from 8 to 64 (ablation) | Handle-based deformation · CUB | mIoU: 0.64 -> 0.711 | Helped |
| Increase the number of deformation handles from 8 to 64 (ablation) | Handle-based deformation · CUB | PCK: 84.6 -> 91.5 | Helped |
| Use 8 deformation handles instead of 64 (ablation) | Handle-based deformation · CUB | PCK: 91.5 -> 84.6 | Hurt |
Closest papers by meaning
- Learning Implicit Representation for Reconstructing Articulated Objects (ICLR 2024 · similarity 0.91)
- Articulation in Prime: Primitive-Based Articulated Object Understanding from a Single Casual Video (NeurIPS 2026 · similarity 0.89)
- S3O: A Dual-Phase Approach for Reconstructing Dynamic Shape and Skeleton of Articulated Objects from Single Monocular Video (ICML 2024 · similarity 0.89)
- REACTO: Reconstructing Articulated Objects from a Single Video (CVPR 2024 · similarity 0.89)
- VideoMDM: Towards 3D Human Motion Generation From 2D Supervision (NeurIPS 2026 · similarity 0.88)
- ArticulatedGS: Self-supervised Digital Twin Modeling of Articulated Objects using 3D Gaussian Splatting (CVPR 2025 · similarity 0.88)
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.