Research map / 3D vision / 3D generation
Extending Layered Models to 3D Motion
Dong Lao, Ganesh Sundaramoorthi
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 coarse-to-fine refinement on overlapping frame batches (final stage) to the initialization (ablation) | flattened layered model · DAVIS 2016 | J mean: 0.491 -> 0.683 | Improved |
| add the 1st coarse-to-fine stage (ablation) | flattened layered model · DAVIS 2016 | J mean: 0.491 -> 0.571 | Improved |
| add the 2nd coarse-to-fine stage (ablation) | flattened layered model · DAVIS 2016 | J mean: 0.571 -> 0.644 | Improved |
| add the 3rd coarse-to-fine stage (ablation) | flattened layered model · DAVIS 2016 | J mean: 0.644 -> 0.673 | Improved |
| add the final coarse-to-fine step after the 3rd stage (ablation) | flattened layered model · DAVIS 2016 | J mean: 0.673 -> 0.683 | Improved |
| coarse-to-fine refinement overall (ablation) | flattened layered model · DAVIS 2016 | F mean: 0.509 -> 0.672 | Improved |
| coarse-to-fine refinement overall (ablation) | flattened layered model · DAVIS 2016 | J recall: 0.575 -> 0.777 | Improved |
| coarse-to-fine refinement overall (ablation) | flattened layered model · DAVIS 2016 | J decay: 0.097 -> 0.069 | Improved |
| add shape evolution of the flattened representation (self-occlusion handling) versus layered method [16] | layered segmentation · DAVIS 2016 | J mean: 0.615 -> 0.683 | Improved |
| add shape evolution of the flattened representation versus [16] | layered segmentation · DAVIS 2016 | F mean: 0.593 -> 0.672 | Improved |
| add shape evolution versus [16] | layered segmentation · DAVIS 2016 | J decay: 0.041 -> 0.069 | Hurt |
| model self-occlusion with flattened 3D representations versus layered method [19] | layered segmentation · DAVIS 2016 | J mean: 0.514 -> 0.683 | Improved |
Closest papers by meaning
- Layered Motion Fusion: Lifting Motion Segmentation to 3D in Egocentric Videos (CVPR 2025 · similarity 0.88)
- Segmenting Moving Objects via an Object-Centric Layered Representation (NeurIPS 2022 · similarity 0.88)
- The Visual Centrifuge: Model-Free Layered Video Representations (CVPR 2019 · similarity 0.88)
- Slice3D: Multi-Slice Occlusion-Revealing Single View 3D Reconstruction (CVPR 2024 · similarity 0.85)
- Motion4D: Learning 3D-Consistent Motion and Semantics for 4D Scene Understanding (NeurIPS 2025 · similarity 0.85)
- Using Diffusion Priors for Video Amodal Segmentation (CVPR 2025 · 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.
