Research map / 3D vision / Face and shape
Learning Neural Deformation Representation for 4D Dynamic Shape Generation
Gyojin Han, Jiwan Hur, Jaehyun Choi, Junmo Kim
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 separated neural deformation representation with a diffusion model on latents instead of HyperDiffusion 4D occupancy fields | Ours (Unconditional) · DeformingThings4D | COV (%): 41.8 -> 52.8 | Improved |
| Use per-vertex animation (mesh extracted once) instead of per-frame marching cubes | Ours (Unconditional) · DeformingThings4D | mesh extraction time (s): 40.4 -> 8.1 | Improved |
| Use proposed method instead of HyperDiffusion for quality | Ours (Unconditional) · DeformingThings4D | MMD: 19.6 -> 16.4 | Improved |
| Use proposed method instead of HyperDiffusion for distribution similarity | Ours (Unconditional) · DeformingThings4D | 1-NNA (%): 64.1 -> 61.9 | Improved |
| Add point-cloud conditioning to the diffusion model | Ours · DeformingThings4D | MMD: 16.4 -> 15.6 | Improved |
| Add point-cloud conditioning | Ours · DeformingThings4D | 1-NNA (%): 61.9 -> 62.5 | No clear effect |
| Replace CaDeX deformation module with skinning-weight deformation representation (ablation) | CaDeX · D-FAUST 4D completion (with Corr supervision) | Corr: 0.095 -> 0.093 | Improved |
| Replace CaDeX NVP deformation module with ours (ablation) | CaDeX · D-FAUST 4D completion (with Corr supervision) | IoU (%): 78.1 -> 78.1 | No clear effect |
Closest papers by meaning
- DNF: Unconditional 4D Generation with Dictionary-based Neural Fields (CVPR 2025 · similarity 0.92)
- Dynamic Neural Surfaces for Elastic 4D Shape Representation and Analysis (CVPR 2025 · similarity 0.89)
- Implicit-ARAP: Efficient Handle-Guided Neural Field Deformation via Local Patch Meshing (NeurIPS 2025 · similarity 0.87)
- Motion2VecSets: 4D Latent Vector Set Diffusion for Non-rigid Shape Reconstruction and Tracking (CVPR 2024 · similarity 0.87)
- Real-time Photorealistic Dynamic Scene Representation and Rendering with 4D Gaussian Splatting (ICLR 2024 · similarity 0.87)
- 4Deform: Neural Surface Deformation for Robust Shape Interpolation (CVPR 2025 · similarity 0.87)
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.
