Research map / 3D vision / Face and shape
Combining Local-Physical and Global-Statistical Models for Sequential Deformable Shape from Motion
Antonio Agudo, Francesc Moreno-Noguer
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 local physical constraints to global-only model (ablation) | GSMM · Mocap Drink | eps3D [%]: 4.48 -> 1.92 | Improved |
| add local physical constraints to global-only model (ablation) | GSMM · Mocap Stretch | eps3D [%]: 6.52 -> 5.65 | Improved |
| add local physical constraints to global-only model (ablation) | GSMM · Mocap Yoga | eps3D [%]: 7.89 -> 6.65 | Improved |
| add global statistical constraints to local-only model (ablation) | LSMM · Mocap Face | eps3D [%]: 4.49 -> 4.33 | Improved |
| add global statistical constraints to local-only model (ablation) | LSMM · Mocap Shark | eps3D [%]: 6.99 -> 6.73 | Hurt |
| add global constraints to local-only model (ablation) | LSMM · Mocap Yoga | eps3D [%]: 6.65 -> 6.65 | No clear effect |
| use GLSMM instead of sequential bundle adjustment | SBA · Mocap Stretch | eps3D [%]: 17.61 -> 5.65 | Improved |
| use sequential GLSMM instead of batch methods | batch KSTA · Back sequence | eps3D [%]: 9.33 -> 7.63 | Improved |
| use sequential GLSMM instead of batch KSTA | KSTA · Mocap Drink | eps3D [%]: 0.94 -> 1.92 | Hurt |
| perturb optimization weights by -25% (ablation) | GLSMM · Mocap Shark | eps3D [%]: 6.73 -> 6.26 | No clear effect |
| perturb optimization weights by +25% (ablation) | GLSMM · Mocap Drink | eps3D [%]: 1.92 -> 2.25 | Hurt |
| initialize force parameter to zero (ablation) | GLSMM · Mocap sequences | convergence time: Stretch (36.45%) | Hurt |
Closest papers by meaning
- Robust Spatio-Temporal Clustering and Reconstruction of Multiple Deformable Bodies (TPAMI 2018 · similarity 0.85)
- DeformMaster: An Interactive Physics-Neural World Model for Deformable Objects from Videos (NeurIPS 2026 · similarity 0.85)
- NPMs: Neural Parametric Models for 3D Deformable Shapes (ICCV 2021 · similarity 0.84)
- Force-Based Representation for Non-Rigid Shape and Elastic Model Estimation (TPAMI 2017 · similarity 0.84)
- Deform360: A Massive Multi-view Visuotactile Dataset for Deformable World Models (ECCV 2026 · similarity 0.84)
- Non-Rigid Structure-from-Motion: Temporally-Smooth Procrustean Alignment and Spatially-Variant Deformation Modeling (CVPR 2024 · similarity 0.83)
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.
