# colmap/colmap

COLMAP - Structure-from-Motion and Multi-View Stereo

Repository: https://github.com/colmap/colmap
Canonical: https://ross.abutalabs.com/products/colmap
Homepage: https://colmap.github.io/
Language: C++
License: NOASSERTION
License Family: other
Topics: structure-from-motion, multi-view-stereo, reconstruction, geometry, computer-vision
Last push: 2026-08-26T17:21:33+00:00

## Health v2 (maintenance only)
Score: 98/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 95, longevity 100
- inputs: {"age_days": 4401, "days_push": 7, "days_rel": 39, "gap_med": 19, "n_releases_24m": 18}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 12564, forks 2127 (observed 2026-08-28T04:10:59.316297+00:00)

## What it is
COLMAP is a general-purpose Structure-from-Motion (SfM) and Multi-View Stereo (MVS) pipeline for reconstructing 3D models from ordered or unordered image collections. It provides both a graphical interface and a command-line interface, plus Python bindings (PyCOLMAP), and is written in C++.

## Use cases
- reconstruct 3D models from a set of photos
- estimate camera poses from unordered image collections
- generate dense point clouds and meshes with multi-view stereo
- run incremental structure-from-motion pipelines from the command line
- use Python bindings to script 3D reconstruction workflows
- prepare camera poses and sparse point clouds for Gaussian splatting or NeRF
- support multi-camera rigs and various camera models in reconstruction

## When to choose
- you need a robust, battle-tested SfM/MVS pipeline with both GUI and CLI
- you want to reconstruct 3D scenes from photos without writing your own algorithms
- you need Python bindings to integrate reconstruction into ML pipelines
- you require support for many camera models and multi-camera rigs

## When to avoid
- you need real-time SLAM on video streams rather than offline reconstruction
- you want a lightweight library for a single geometric operation only
- you need a fully automated cloud service rather than self-hosted software
- your project requires a permissive license and the BSD terms here are unacceptable

## Facets
- artifact type: application
- maturity: active
- function: computer-vision, image-processing, graphics, cli, gui, sdk
- domain: computer-vision, graphics
- platform: windows, python, cpp, cross-platform
- tags: structure-from-motion, multi-view-stereo, photogrammetry, 3d-reconstruction, point-cloud, camera-pose-estimation, pycolmap, cuda, linux, macos, docker

## Member repositories
- colmap/colmap (main) score 98

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:59.316297+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 2026-08-29T17:13:57.966776+00:00, confidence not recorded.
  - readme: https://github.com/colmap/colmap (fetched 2026-08-28T04:10:59.316297+00:00, sha 5cda4a1c0957)
  - homepage: https://colmap.github.io/ (fetched 2026-08-29T08:10:29.559778+00:00, sha 50fb25e0d78b)
- Data as of 2026-08-30T08:39:29.467469+00:00.