Research map / Video / Video segmentation and tracking
Omnimatte3D: Associating Objects and Their Effects in Unconstrained Monocular Video
Mohammed Suhail, Erika Lu, Zhengqi Li, Noah Snavely, Leonid Sigal, Forrester Cole
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 |
|---|---|---|---|
| Use 3D background layers with multi-view projection consistency | Omnimatte3D · DAVIS videos with camera parallax | background and foreground layer decomposition quality | Improved |
| Use 3D background layers with multi-view projection consistency | Omnimatte3D · Snowboard | background inpainting quality | Improved |
| Use 3D background layers with multi-view projection consistency | Omnimatte3D · Scooter-gray | background and foreground layer decomposition quality | Improved |
| Use 3D background layers with multi-view projection consistency | Omnimatte3D · Longboard | foreground disentanglement quality | Improved |
| Select training frames using the proposed overlap heuristic instead of random sampling (ablation) | Omnimatte3D · Hike | shadow assignment to the foreground layer | Improved |
| Add projection consistency loss (ablation) | Omnimatte3D · Car scene in Figure 9 | background inpainting quality | Improved |
| Add projection consistency loss (ablation) | Omnimatte3D · Car scene in Figure 9 | shadow capture in the foreground layer | Improved |
| Transfer background detail from nearby frames instead of using only the current frame (ablation) | Omnimatte3D · Bicyclist scene | background detail quality | Improved |
Closest papers by meaning
- Generative Omnimatte: Learning to Decompose Video into Layers (CVPR 2025 · similarity 0.90)
- MonoSplat: Generalizable 3D Gaussian Splatting from Monocular Depth Foundation Models (CVPR 2025 · similarity 0.86)
- DreamScene4D: Dynamic Multi-Object Scene Generation from Monocular Videos (NeurIPS 2024 · similarity 0.86)
- Reconstruct, Inpaint, Test-Time Finetune: Dynamic Novel-view Synthesis from Monocular Videos (NeurIPS 2025 · similarity 0.85)
- OmniLocalRF: Omnidirectional Local Radiance Fields from Dynamic Videos (CVPR 2024 · similarity 0.85)
- ODGS: 3D Scene Reconstruction from Omnidirectional Images with 3D Gaussian Splattings (NeurIPS 2024 · similarity 0.85)
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.
