Research map / 3D vision / Face and shape
A Level Set Theory for Neural Implicit Evolution under Explicit Flows
Ishit Mehta, Manmohan Chandraker, Ravi Ramamoorthi
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 |
|---|---|---|---|
| evolve implicit surface with level-set objective instead of differentiable rasterization-free single-step MeshSDF update | parametric level-set inverse rendering · Genus6 inverse rendering | PSNR (dB): 25.67 -> 32.94 | Improved |
| use explicit differentiable renderer on level-set instead of mesh-only LSIG (topology can change) | inverse rendering · Bunny (genus-0) | Chamfer distance (x1e-3): 0.06 vs 0.18 | No clear effect |
| use explicit differentiable renderer on level-set instead of mesh-only LSIG | inverse rendering · Rind and VSphere | Chamfer distance (x1e-3): 3.94 -> 5.56; 4.78 -> 3.71 | Hurt |
| use level-set method instead of LSIG mesh optimization | inverse rendering · Kangaroo | PSNR (dB): 25.52 -> 26.48 | Improved |
| add smoothness regularization term to flow field (ablation) | inverse rendering · inverse rendering shapes | quantitative error | No clear effect |
| evolve with level-set objective instead of Laplacian of MLP (NFGP) | surface smoothing · smoothing computer graphics models | qualitative smoothness | Improved |
| minimize level-set objective J to convergence instead of single MeshSDF step | mean-curvature flow · genus-0 surface to sphere | qualitative flow | Improved |
| use level-set editing instead of NFGP | user-defined shape editing · cylinder rotation | editing time: 10 hr -> 10 min | Improved |
Closest papers by meaning
- Implicit Neural Surface Deformation with Explicit Velocity Fields (ICLR 2025 · similarity 0.91)
- Neural Implicit Surface Evolution (ICCV 2023 · similarity 0.91)
- Implicit-ARAP: Efficient Handle-Guided Neural Field Deformation via Local Patch Meshing (NeurIPS 2025 · similarity 0.89)
- 4Deform: Neural Surface Deformation for Robust Shape Interpolation (CVPR 2025 · similarity 0.87)
- Flatten Anything: Unsupervised Neural Surface Parameterization (NeurIPS 2024 · similarity 0.87)
- Small Steps and Level Sets: Fitting Neural Surface Models with Point Guidance (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.
