Research map / 3D vision / Face and shape
Amodal3R: Amodal 3D Reconstruction from Occluded 2D Images
Tianhao Wu, Chuanxia Zheng, Frank Guan, Andrea Vedaldi, Tat-Jen Cham
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 |
|---|---|---|---|
| replace naive token concatenation with mask-weighted attention plus occlusion-aware layer (full model) (ablation) | Amodal3R · GSO single-view | FID: 31.96 -> 30.64 | Improved |
| replace naive token concatenation with mask-weighted attention plus occlusion-aware layer (full model) (ablation) | Amodal3R · GSO single-view | KID (%): 0.49 -> 0.35 | Improved |
| add occlusion-aware layer to mask-weighted attention (ablation) | Amodal3R · GSO single-view | COV (%): 36.90 -> 39.61 | Improved |
| add occlusion-aware layer to mask-weighted attention (ablation) | Amodal3R · GSO single-view | MMD: 3.69 -> 3.62 | Improved |
| add mask-weighted attention to occlusion-aware layer (ablation) | Amodal3R · GSO single-view | KID (%): 0.57 -> 0.35 | Improved |
| add mask-weighted attention to occlusion-aware layer (ablation) | Amodal3R · GSO single-view | COV (%): 40.19 -> 39.61 | Hurt |
| use 4 occluded views instead of 1 view | Amodal3R · GSO | FID: 30.64 -> 26.27 | Improved |
| use 4 independently pix2gestalt-completed views instead of 1 completed view | TRELLIS · GSO | FID: 58.82 -> 65.69 | Hurt |
| use 4 views completed by pix2gestalt+MV instead of 1 completed view | TRELLIS · GSO | KID (%): 5.87 -> 1.85 | Improved |
| use single-stage Amodal3R instead of pix2gestalt + TRELLIS | TRELLIS · Toys4K single-view | FID: 43.05 -> 23.45 | Improved |
Later papers that compared with this method
3 later accepted papers report Amodal3R as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.
- GENA3D: Generative Amodal 3D Modeling by Bridging 2D Priors and 3D Coherence (ECCV 2026 · better on 7 of 7 shared results)
- Physically Grounded 3D Generative Reconstruction under Hand Occlusion using Proprioception and Multi-Contact Touch (ECCV 2026 · better on 6 of 6 shared results)
- SceneMaker: Open-set 3D Scene Generation with Decoupled De-occlusion and Pose Estimation Model (CVPR 2026 · better on 3 of 3 shared results)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Open-World Amodal Appearance Completion (CVPR 2025 · similarity 0.89)
- In-Hand 3D Object Reconstruction from a Monocular RGB Video (AAAI 2024 · similarity 0.88)
- Amodal Scene Analysis via Holistic Occlusion Relation Inference and Generative Mask Completion (AAAI 2024 · similarity 0.88)
- pix2gestalt: Amodal Segmentation by Synthesizing Wholes (CVPR 2024 · similarity 0.88)
- Slice3D: Multi-Slice Occlusion-Revealing Single View 3D Reconstruction (CVPR 2024 · similarity 0.87)
- Using Diffusion Priors for Video Amodal Segmentation (CVPR 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.
