Research map / 3D vision / Face and shape
NeAT: Learning Neural Implicit Surfaces With Arbitrary Topologies From Multi-View Images
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 NeAT instead of the strongest listed watertight reconstruction baseline | DTU reconstruction · DTU scans average | Chamfer Distance: 0.741 -> 0.749 | Hurt |
| Use NeAT instead of NeuS | Open-surface reconstruction · Deep Fashion 3D average | Chamfer Distance (× 10^-3): 6.847 -> 4.315 | Helped |
| Use NeAT instead of NeuS | Open-surface reconstruction · Multi-Garment Net average | Chamfer Distance (× 10^-3): 10.740 -> 9.176 | Helped |
| Use the NeAT autoencoder instead of NeuS | Single-view reconstruction · Deep Fashion 3D dress test set | Chamfer Distance: 0.0778 -> 0.0771 | Helped |
| Use the NeAT autoencoder instead of DVR | Single-view reconstruction · Deep Fashion 3D dress test set | Chamfer Distance: 0.0789 -> 0.0771 | Helped |
| Add validity binary-cross-entropy regularization (ablation) | NeAT · Open-surface reconstruction | Validity quality | Helped |
| Add validity sparsity regularization (ablation) | NeAT · Open-surface reconstruction | Redundant surface formation | Helped |
| Increase the number of input views (ablation) | NeAT · Multi-view reconstruction | Reconstruction quality | Helped |
Closest papers by meaning
- Towards Unbiased Volume Rendering of Neural Implicit Surfaces With Geometry Priors (CVPR 2023 · similarity 0.92)
- NeuralUDF: Learning Unsigned Distance Fields for Multi-View Reconstruction of Surfaces With Arbitrary Topologies (CVPR 2023 · similarity 0.92)
- Geo-Neus: Geometry-Consistent Neural Implicit Surfaces Learning for Multi-view Reconstruction (NeurIPS 2022 · similarity 0.92)
- Voxurf: Voxel-based Efficient and Accurate Neural Surface Reconstruction (ICLR 2023 · similarity 0.91)
- SuperNormal: Neural Surface Reconstruction via Multi-View Normal Integration (CVPR 2024 · similarity 0.89)
- Neural Signed Distance Function Inference through Splatting 3D Gaussians Pulled on Zero-Level Set (NeurIPS 2024 · similarity 0.89)
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.