Research map / 3D vision / Depth and stereo
Amodal Depth Anything: Amodal Depth Estimation in the Wild
Zhenyu Li, Mykola Lavreniuk, Jian Shi, Shariq Farooq Bhat, Peter Wonka
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 observation depth and amodal mask guidance (ablation) | Amodal-DAV2-S · ADIW validation (Amodal-DAV2-S ablation) | RMSE x100: 7.549 -> 3.682 | Improved |
| add amodal mask guidance on top of observation depth (ablation) | Amodal-DAV2-S · ADIW validation (Amodal-DAV2-S ablation) | RMSE x100: 4.369 -> 3.682 | Improved |
| add scale-and-shift alignment at inference (ablation) | Amodal-DAV2-S · ADIW validation (Amodal-DAV2-S ablation) | RMSE x100: 3.682 -> 3.878 | Hurt |
| supervise whole object depth instead of invisible region only (ablation) | Amodal-DAV2-S · ADIW validation (Amodal-DAV2-S ablation) | RMSE x100: 3.845 -> 3.682 | Improved |
| use silog loss instead of scale-shift-invariant loss with alignment (ablation) | Amodal-DAV2-S · ADIW validation (Amodal-DAV2-S ablation) | RMSE x100: 4.015 -> 3.682 | Improved |
| add observation depth and amodal mask guidance (ablation) | Amodal-DepthFM · ADIW validation (Amodal-DepthFM ablation) | RMSE x100: 10.211 -> 4.645 | Improved |
| add amodal mask guidance | Amodal-DepthFM · ADIW validation (Amodal-DepthFM ablation) | RMSE x100: 5.283 -> 5.410 | Improved |
| add scale-and-shift alignment at inference (ablation) | Amodal-DepthFM · ADIW validation (Amodal-DepthFM ablation) | RMSE x100: 5.410 -> 4.645 | Improved |
| supervise whole object depth instead of invisible region only (ablation) | Amodal-DepthFM · ADIW validation (Amodal-DepthFM ablation) | RMSE x100: 5.636 -> 4.645 | Improved |
| use flow matching objective instead of RMSE scene loss (ablation) | Amodal-DepthFM · ADIW validation (Amodal-DepthFM ablation) | RMSE x100: 4.608 -> 4.645 | No clear effect |
| use Amodal-DAV2-L instead of Amodal-DepthFM (aligned) | Amodal-DAV2 · ADIW validation (overall) | RMSE x100: 4.645 -> 3.418 | Improved |
| scale model from ViT-S to ViT-L | Amodal-DAV2 · ADIW validation (overall) | RMSE x100: 3.682 -> 3.418 | Improved |
Closest papers by meaning
- Amodal Ground Truth and Completion in the Wild (CVPR 2024 · similarity 0.89)
- Open-World Amodal Appearance Completion (CVPR 2025 · similarity 0.88)
- Using Diffusion Priors for Video Amodal Segmentation (CVPR 2025 · similarity 0.86)
- A Simple yet Universal Framework for Depth Completion (NeurIPS 2024 · similarity 0.86)
- RSA: Resolving Scale Ambiguities in Monocular Depth Estimators through Language Descriptions (NeurIPS 2024 · similarity 0.86)
- Segment Anything, Even Occluded (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.
