# NVlabs/neuralangelo

Official implementation of "Neuralangelo: High-Fidelity Neural Surface Reconstruction" (CVPR 2023)

Repository: https://github.com/NVlabs/neuralangelo
Canonical: https://ross.abutalabs.com/products/neuralangelo
Homepage: https://research.nvidia.com/labs/dir/neuralangelo/
Language: Python
License: NOASSERTION
License Family: other
Topics: 3d-reconstruction, computer-graphics, computer-vision, neural-rendering, neural-representation
Last push: 2024-04-14T21:29:12+00:00

## Health v2 (maintenance only)
Score: 29/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 83
- inputs: {"age_days": 1167, "days_push": 871, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 4615, forks 404 (observed 2026-08-28T04:08:55.114933+00:00)

## What it is
Official PyTorch implementation of Neuralangelo, a CVPR 2023 method for high-fidelity neural surface reconstruction from multi-view images. It builds on NVIDIA's Imaginaire library and trains neural signed distance functions with multi-GPU support.

## Use cases
- reconstruct 3d surfaces from video frames
- high-fidelity neural surface reconstruction from images
- generate 3d meshes from multi-view photos
- research neural implicit surface modeling
- reproduce cvpr 2023 neuralangelo paper results

## When to choose
- you need state-of-the-art neural surface reconstruction with fine geometric detail
- you have known camera poses and want to train a neural SDF on multi-view data
- you want the official reference implementation for research reproduction

## When to avoid
- you need a quick photogrammetry pipeline without GPU training
- you lack CUDA GPUs or multi-GPU infrastructure
- you want a turnkey 3D scanning app rather than a research codebase

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, computer-vision, graphics, simulation
- domain: computer-vision, graphics, deep-learning
- platform: python
- tags: neural-rendering, 3d-reconstruction, neural-surface-reconstruction, sdf, nerf, cvpr-2023, research-code, multi-gpu-training, gpu, docker, linux

## Member repositories
- NVlabs/neuralangelo (main) score 29

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:55.114933+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:19:40.508848+00:00, confidence not recorded.
  - readme: https://github.com/NVlabs/neuralangelo (fetched 2026-08-28T04:08:55.114933+00:00, sha 609129173ac4)
  - homepage: https://research.nvidia.com/labs/dir/neuralangelo/ (fetched 2026-08-29T09:04:33.256892+00:00, sha 8635b92ed92e)
- Data as of 2026-08-30T08:39:29.467469+00:00.