Research map / 3D vision / Face and shape
NeuS: Learning Neural Implicit Surfaces by Volume Rendering for Multi-view Reconstruction
Read the paperCheck 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 |
|---|---|---|---|
| use NeuS unbiased opaque-density weight instead of naive volume-rendering weight w=T*sigma with S-density (ablation) | NeuS · DTU (ablation scene) | Chamfer distance | Helped |
| use NeuS weight instead of direct normalized S-density weight (Eqn 4) (ablation) | NeuS · DTU (ablation scene) | reconstruction quality | Helped |
| add Eikonal regularization (ablation) | NeuS · DTU (ablation scene) | Chamfer distance | No clear effect |
| add geometric initialization (ablation) | NeuS · DTU (ablation scene) | Chamfer distance | No clear effect |
| add Eikonal regularization and geometric initialization (ablation) | NeuS · DTU (ablation scene) | SDF MAE | Helped |
| NeuS vs IDR, without mask supervision for NeuS | NeuS · DTU | mean Chamfer distance: 0.90 (IDR w/ mask) vs 0.84 (NeuS w/o mask) | Mixed |
| replace surface rendering (IDR) with NeuS volume rendering, both with mask | IDR · DTU w/ mask | mean Chamfer distance: 0.90 -> 0.77 | Helped |
| replace NeRF density with NeuS SDF volume rendering, w/o mask | NeRF · DTU w/o mask | mean Chamfer distance: 1.49 -> 0.84 | Helped |
| replace UNISURF occupancy with NeuS SDF representation, w/o mask | UNISURF · DTU w/o mask | mean Chamfer distance: 1.02 -> 0.84 | Helped |
Closest papers by meaning
- SC-NeuS: Consistent Neural Surface Reconstruction from Sparse and Noisy Views (AAAI 2024 · similarity 0.93)
- From Transparent to Opaque: Rethinking Neural Implicit Surfaces with $\alpha$-NeuS (NeurIPS 2024 · similarity 0.90)
- NeuSurf: On-Surface Priors for Neural Surface Reconstruction from Sparse Input Views (AAAI 2024 · similarity 0.90)
- NeuRodin: A Two-stage Framework for High-Fidelity Neural Surface Reconstruction (NeurIPS 2024 · similarity 0.89)
- UniSDF: Unifying Neural Representations for High-Fidelity 3D Reconstruction of Complex Scenes with Reflections (NeurIPS 2024 · similarity 0.89)
- ViiNeuS: Volumetric Initialization for Implicit Neural Surface Reconstruction of Urban Scenes with Limited Image Overlap (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.