Research map / 3D vision / Face and shape
Improving Neural Implicit Surfaces Geometry With Patch Warping
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 patch warping with volumetric rendering and occlusion masking | VolSDF · DTU | mean chamfer distance: 0.86 -> 0.68 | Helped |
| Add patch warping with volumetric rendering and occlusion masking | VolSDF · EPFL Center | chamfer distance: 3.78 -> 1.97 | Helped |
| Add patch warping with volumetric rendering and occlusion masking | VolSDF · EPFL Full | chamfer distance: 13.25 -> 8.32 | Helped |
| Use NeuralWarp instead of COLMAP | EPFL Full · EPFL Full | chamfer distance: 7.21 -> 8.32 | Hurt |
| Use NeuralWarp instead of COLMAP | EPFL Center · EPFL Center | chamfer distance: 2.38 -> 1.97 | Helped |
| Add pixel warping to volumetric rendering (ablation) | VolSDF · DTU | mean chamfer distance: 0.85 -> 0.83 | Helped |
| Replace pixel warping with patch warping (ablation) | Pixel · DTU | mean chamfer distance: 0.83 -> 0.68 | Helped |
| Add occlusion masking to patch warping (ablation) | Patch no occ. · DTU | mean chamfer distance: 0.74 -> 0.68 | Helped |
| Add volumetric rendering loss to patch warping (ablation) | Patch no vol. · DTU | mean chamfer distance: 0.74 -> 0.68 | Helped |
| Remove the projection mask (ablation) | NeuralWarp · DTU | reconstruction quality | Hurt |
Closest papers by meaning
- Differentiable Neural Surface Refinement for Modeling Transparent Objects (CVPR 2024 · similarity 0.88)
- NeuSurf: On-Surface Priors for Neural Surface Reconstruction from Sparse Input Views (AAAI 2024 · similarity 0.88)
- From Transparent to Opaque: Rethinking Neural Implicit Surfaces with $\alpha$-NeuS (NeurIPS 2024 · similarity 0.87)
- Learning 3D Equivariant Implicit Function with Patch-Level Pose-Invariant Representation (NeurIPS 2024 · similarity 0.87)
- GeoSVR: Taming Sparse Voxels for Geometrically Accurate Surface Reconstruction (NeurIPS 2025 · similarity 0.86)
- UFORecon: Generalizable Sparse-View Surface Reconstruction from Arbitrary and Unfavorable Sets (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.