Volume Rendering with Metric--Phase Fields for Thin Structures and Sign-Ambiguous Geometry
Neural implicit surface methods that couple volume rendering with signed distance functions (SDFs), such as VolSDF and NeuS, achieve high-quality multi-view reconstruction for watertight objects.
ProofPaper ↗
Key points
- We propose PhaseSurf, a neural volume rendering framework built on Metric-Phase Fields (MPFs).
- MPFs decompose geometry into a nonnegative metric field capturing unsigned proximity and a continuous phase field providing a soft sign cue.
- We further introduce MPF-specific training losses that enforce metric consistency, stabilize phase transitions near the interface, and suppress spurious geometry in free space.
- Experiments on multi-view datasets with thin structures and sign-ambiguous geometry show that PhaseSurf produces more stable reconstructions and higher-quality geometry than UDF-based methods and recent 3DGS pipelines, while avoiding artifacts typical of SDF-based approaches.
Sources (1)
- [1]Volume Rendering with Metric--Phase Fields for Thin Structures and Sign-Ambiguous GeometryarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 9, 06:40 AM
Neural implicit surface methods that couple volume rendering with signed distance functions (SDFs), such as VolSDF and NeuS, achieve high-quality multi-view reconstruction for watertight objects.
We propose PhaseSurf, a neural volume rendering framework built on Metric-Phase Fields (MPFs).
Extractive summary: sentences quoted from the sources.