Research map / 3D vision / Face and shape
DeepDeform: Learning Non-Rigid RGB-D Reconstruction With Semi-Supervised Data
Aljaz Bozic, Michael Zollhofer, Christian Theobalt, Matthias Niessner
Check 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 learned non-rigid RGB-D matching instead of FlowNet-2.0 | Non-rigid matching · Non-rigid matching test set | 2D error: 27.32 -> 19.56 | Improved |
| use semi-supervised dense alignment training data instead of self-supervised data (ablation) | DeepDeform matching network · Non-rigid matching test set | 2D error: 33.34 -> 19.56 | Improved |
| use semi-supervised dense alignment training data instead of sparse-only annotations (ablation) | DeepDeform matching network · Non-rigid matching test set | 2D error: 31.42 -> 19.56 | Improved |
| add deformation-aware sample reweighting during training (ablation) | DeepDeform matching network · Non-rigid matching test set | 2D error: 23.72 -> 19.56 | Improved |
| train on a reduced 12.5% subset of the training data (ablation) | DeepDeform matching network · Non-rigid matching test set | 2D error: 19.56 -> 78.82 | Hurt |
| train on a reduced 25.0% subset of the training data (ablation) | DeepDeform matching network · Non-rigid matching test set | 2D error: 19.56 -> 58.28 | Hurt |
| train on a reduced 50.0% subset of the training data (ablation) | DeepDeform matching network · Non-rigid matching test set | 2D error: 19.56 -> 45.43 | Hurt |
| add visibility-based correspondence pruning (ablation) | DeepDeform reconstruction · Non-rigid reconstruction test sequences | Deformation error (cm): 3.94 -> 3.70 | Improved |
| add depth-based correspondence filtering (ablation) | DeepDeform reconstruction with visibility pruning · Non-rigid reconstruction test sequences | Deformation error (cm): 3.70 -> 3.28 | Improved |
Later papers that compared with this method
2 later accepted papers report DeepDeform as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- OcclusionFusion: Occlusion-Aware Motion Estimation for Real-Time Dynamic 3D Reconstruction (CVPR 2022 · better on 2 of 2 shared results)
- Neural Non-Rigid Tracking (NeurIPS 2020 · better on 2 of 2 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
- Neural Dense Non-Rigid Structure from Motion with Latent Space Constraints (ECCV 2020 · similarity 0.89)
- TFS-NeRF: Template-Free NeRF for Semantic 3D Reconstruction of Dynamic Scene (NeurIPS 2024 · similarity 0.88)
- Deep Non-Rigid Structure-from-Motion Revisited: Canonicalization and Sequence Modeling (AAAI 2025 · similarity 0.86)
- Motion2VecSets: 4D Latent Vector Set Diffusion for Non-rigid Shape Reconstruction and Tracking (CVPR 2024 · similarity 0.86)
- MorpheuS: Neural Dynamic 360deg Surface Reconstruction from Monocular RGB-D Video (CVPR 2024 · similarity 0.86)
- Neural Parametric Gaussians for Monocular Non-Rigid Object Reconstruction (CVPR 2024 · 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.
