# niessner/BundleFusion

[Siggraph 2017] BundleFusion: Real-time Globally Consistent 3D Reconstruction using Online Surface Re-integration

Repository: https://github.com/niessner/BundleFusion
Canonical: https://ross.abutalabs.com/products/bundlefusion
Homepage: http://graphics.stanford.edu/projects/bundlefusion/
Language: C++
License: NOASSERTION
License Family: other
Topics: reconstruction, computervision, 3d-reconstruction, cvpr, siggraph, kinect
Last push: 2020-05-29T12:36:00+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 3320, "days_push": 2287, "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 1588, forks 356 (observed 2026-08-28T04:05:07.970182+00:00)

## What it is
BundleFusion is a real-time, globally consistent 3D reconstruction system for RGB-D input, published at SIGGRAPH 2017. It estimates globally optimized camera poses online and re-integrates surfaces in real time, supporting robust tracking and relocalization.

## Use cases
- reconstruct 3D models from RGB-D camera streams in real time
- scan large-scale indoor scenes with globally consistent geometry
- recover from tracking failures during 3D scanning
- build mixed reality or robotics prototypes needing live 3D maps
- reproduce SIGGRAPH 2017 research results on 3D reconstruction

## When to choose
- you need real-time RGB-D 3D reconstruction with global consistency
- you have a Windows machine with an NVIDIA GPU and CUDA
- you want a research baseline for SLAM or surface re-integration

## When to avoid
- you need a maintained cross-platform library or active community support
- you target Linux or macOS
- you need commercial-use licensing

## Facets
- artifact type: application
- maturity: maintenance
- function: computer-vision, graphics, simulation
- domain: computer-vision, graphics, robotics
- platform: windows, cpp
- tags: 3d-reconstruction, rgb-d, slam, cuda, kinect, real-time, research-code, gpu

## Member repositories
- niessner/BundleFusion (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:07.970182+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 2026-08-30T03:55:04.431575+00:00, confidence not recorded.
  - readme: https://github.com/niessner/BundleFusion (fetched 2026-08-28T04:05:07.970182+00:00, sha b160a416d3dd)
  - homepage: http://graphics.stanford.edu/projects/bundlefusion/ (fetched 2026-08-29T11:25:43.185389+00:00, sha 52c7a29d324f)
- Data as of 2026-08-30T08:39:29.467469+00:00.