snap-research/NeROIC

Health v2 · maintenance only

Flags: no_releases no_license

Adoption not part of the score

1012 stars · 132 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

Official PyTorch implementation of NeROIC, a neural method for capturing 3D object geometry and material from online image collections and rendering novel views with relighting. It trains geometry and rendering networks in stages from multi-view images.

Use cases

• reconstruct 3d objects from photos
• novel view synthesis of an object
• relight a captured 3d object
• decompose materials from images
• neural object capture from online images

When to choose

• you need research-grade neural 3D object capture from multi-view images
• you want to reproduce the NeROIC paper results
• you need relighting or material decomposition of captured objects

When to avoid

• you need a production-ready 3D scanning pipeline
• you lack GPU resources for multi-stage training
• you need Windows or macOS support without adaptation

Facets

library · maturity maintenance

machine-learning deep-learning image-processing computer-vision graphics simulation computer-vision graphics deep-learning machine-learning python neural-rendering 3d-reconstruction nerf novel-view-synthesis relighting research-code pytorch linux gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="snap-research/NeROIC")