Research map / 3D vision / Face and shape
Iso-Points: Optimizing Neural Implicit Surfaces With Hybrid Representations
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 |
|---|---|---|---|
| Replace ray-tracing samples with loss-based iso-point importance sampling | SIREN multi-view reconstruction · Sketchfab (10 models) | CD · 10⁻⁴ (position): 17.24 -> 1.71 | Helped |
| Replace ray-tracing samples with loss-based iso-point importance sampling | SIREN multi-view reconstruction · Sketchfab (10 models) | CD · 10⁻¹ (normal): 1.51 -> 0.95 | Helped |
| Use uniform iso-points rather than ray-tracing samples | SIREN multi-view reconstruction · Sketchfab (10 models) | CD · 10⁻⁴ (position): 17.24 -> 1.80 | Helped |
| Use curvature-based rather than uniform iso-point sampling (ablation) | SIREN multi-view reconstruction · Sketchfab (10 models) | CD · 10⁻⁴ (position): 1.80 -> 1.83 | Hurt |
| Use curvature-based rather than uniform iso-point sampling (ablation) | SIREN multi-view reconstruction · Sketchfab (10 models) | CD · 10⁻¹ (normal): 1.10 -> 0.99 | Helped |
| Use loss-based rather than uniform iso-point sampling (ablation) | SIREN multi-view reconstruction · Sketchfab (10 models) | CD · 10⁻⁴ (position): 1.80 -> 1.71 | Helped |
| Add iso-point regularization and outlier-aware weighting | SIREN point-cloud reconstruction · DTU-MVS (average) | L1-Chamfer distance: 0.61 -> 0.53 | Helped |
| Use iso-point regularization rather than the vanilla baseline | SIREN point-cloud reconstruction · DTU-MVS model 110 | L1-Chamfer distance: 0.51 -> 0.54 | Hurt |
| Use iso-point regularization rather than screened Poisson | Point-cloud reconstruction · DTU-MVS (average) | L1-Chamfer distance: 0.47 -> 0.53 | Hurt |
Closest papers by meaning
- Neural Unsigned Distance Fields for Implicit Function Learning (NeurIPS 2020 · similarity 0.91)
- Geo-Neus: Geometry-Consistent Neural Implicit Surfaces Learning for Multi-view Reconstruction (NeurIPS 2022 · similarity 0.91)
- NeAT: Learning Neural Implicit Surfaces With Arbitrary Topologies From Multi-View Images (CVPR 2023 · similarity 0.90)
- NeuralIndicator: Implicit Surface Reconstruction from Neural Indicator Priors (ICML 2024 · similarity 0.90)
- LISR: Learning Linear 3D Implicit Surface Representation Using Compactly Supported Radial Basis Functions (AAAI 2024 · similarity 0.90)
- Small Steps and Level Sets: Fitting Neural Surface Models with Point Guidance (CVPR 2024 · similarity 0.88)
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.