Skip to main content

COLMAP bindings

Project description

Python bindings for COLMAP

This repository exposes to Python most capabilities of COLMAP for Structure-from-Motion and Multiview-stereo, such as reconstruction pipelines & objects and geometric estimators.


Wheels for Python 8/9/10 on Linux, macOS 10/11/12 (both Intel and Apple Silicon), and Windows can be installed using pip:

pip install pycolmap

The wheels are automatically built and pushed to PyPI at each release. They are currently not built with CUDA support, which requires building from source.

[Building PyCOLMAP from source - click to expand]
  1. Install COLMAP from source following the official guide. Use COLMAP 3.8 or 3.9.1 for PyCOLMAP 0.4.0 or 0.5.0/0.6.0.

  2. Clone the PyCOLMAP repository:

git clone -b 0.6.0
cd pycolmap
  1. Build:
  • On Linux and macOS:
python -m pip install .
  • On Windows, after installing COLMAP via VCPKG, run in powershell:
py -m pip install . `
    --cmake.define.CMAKE_TOOLCHAIN_FILE="$VCPKG_INSTALLATION_ROOT/scripts/buildsystems/vcpkg.cmake" `

Reconstruction pipeline

PyCOLMAP provides bindings for multiple steps of the standard reconstruction pipeline:

  • extracting and matching SIFT features
  • importing an image folder into a COLMAP database
  • inferring the camera parameters from the EXIF metadata of an image file
  • running two-view geometric verification of matches on a COLMAP database
  • triangulating points into an existing COLMAP model
  • running incremental reconstruction from a COLMAP database
  • dense reconstruction with multi-view stereo

Sparse & Dense reconstruction from a folder of images can be performed with:

output_path: pathlib.Path
image_dir: pathlib.Path

mvs_path = output_path / "mvs"
database_path = output_path / "database.db"

pycolmap.extract_features(database_path, image_dir)
maps = pycolmap.incremental_mapping(database_path, image_dir, output_path)
# dense reconstruction
pycolmap.undistort_images(mvs_path, output_path, image_dir)
pycolmap.patch_match_stereo(mvs_path)  # requires compilation with CUDA
pycolmap.stereo_fusion(mvs_path / "dense.ply", mvs_path)

PyCOLMAP can leverage the GPU for feature extraction, matching, and multi-view stereo if COLMAP was compiled with CUDA support. Similarly, PyCOLMAP can run Delauney Triangulation if COLMAP was compiled with CGAL support. This requires to build the package from source and is not available with the PyPI wheels.

All of the above steps are easily configurable with python dicts which are recursively merged into their respective defaults, for example:

pycolmap.extract_features(database_path, image_dir, sift_options={"max_num_features": 512})
# equivalent to
ops = pycolmap.SiftExtractionOptions()
ops.max_num_features = 512
pycolmap.extract_features(database_path, image_dir, sift_options=ops)

To list available options and their default parameters:


For another example of usage, see or hloc/

Reconstruction object

We can load and manipulate an existing COLMAP 3D reconstruction:

import pycolmap
reconstruction = pycolmap.Reconstruction("path/to/reconstruction/dir")

for image_id, image in reconstruction.images.items():
    print(image_id, image)

for point3D_id, point3D in reconstruction.points3D.items():
    print(point3D_id, point3D)

for camera_id, camera in reconstruction.cameras.items():
    print(camera_id, camera)


The object API mirrors the COLMAP C++ library. The bindings support many other operations, for example:

  • projecting a 3D point into an image with arbitrary camera model:
uv = camera.img_from_cam(image.cam_from_world *
  • aligning two 3D reconstructions by their camera poses:
rec2_from_rec1 = pycolmap.align_reconstructions_via_reprojections(reconstruction1, reconstrution2)
print(rec2_from_rec1.scale, rec2_from_rec1.rotation, rec2_from_rec1.translation)
  • exporting reconstructions to text, PLY, or other formats:
reconstruction.write_text("path/to/new/reconstruction/dir/")  # text format
reconstruction.export_PLY("rec.ply")  # PLY format


We provide robust RANSAC-based estimators for absolute camera pose (single-camera and multi-camera-rig), essential matrix, fundamental matrix, homography, and two-view relative pose for calibrated cameras.

All RANSAC and estimation parameters are exposed as objects that behave similarly as Python dataclasses. The RANSAC options are described in colmap/optim/ransac.h and their default values are:

ransac_options = pycolmap.RANSACOptions(
    max_error=4.0,  # for example the reprojection error in pixels

Absolute pose estimation

For instance, to estimate the absolute pose of a query camera given 2D-3D correspondences:

# Parameters:
# - points2D: Nx2 array; pixel coordinates
# - points3D: Nx3 array; world coordinates
# - camera: pycolmap.Camera
# Optional parameters:
# - estimation_options: dict or pycolmap.AbsolutePoseEstimationOptions
# - refinement_options: dict or pycolmap.AbsolutePoseRefinementOptions
answer = pycolmap.absolute_pose_estimation(points2D, points3D, camera)
# Returns: dictionary of estimation outputs or None if failure

2D and 3D points are passed as Numpy arrays or lists. The options are defined in estimators/ and can be passed as regular (nested) Python dictionaries:

    points2D, points3D, camera,

Absolute Pose Refinement

# Parameters:
# - cam_from_world: pycolmap.Rigid3d, initial pose
# - points2D: Nx2 array; pixel coordinates
# - points3D: Nx3 array; world coordinates
# - inlier_mask: array of N bool; inlier_mask[i] is true if correpondence i is an inlier
# - camera: pycolmap.Camera
# Optional parameters:
# - refinement_options: dict or pycolmap.AbsolutePoseRefinementOptions
answer = pycolmap.pose_refinement(cam_from_world, points2D, points3D, inlier_mask, camera)
# Returns: dictionary of refinement outputs or None if failure

Essential matrix estimation

# Parameters:
# - points1: Nx2 array; 2D pixel coordinates in image 1
# - points2: Nx2 array; 2D pixel coordinates in image 2
# - camera1: pycolmap.Camera of image 1
# - camera2: pycolmap.Camera of image 2
# Optional parameters:
# - options: dict or pycolmap.RANSACOptions (default inlier threshold is 4px)
answer = pycolmap.essential_matrix_estimation(points1, points2, camera1, camera2)
# Returns: dictionary of estimation outputs or None if failure

Fundamental matrix estimation

answer = pycolmap.fundamental_matrix_estimation(
    [options],       # optional dict or pycolmap.RANSACOptions

Homography estimation

answer = pycolmap.homography_matrix_estimation(
    [options],       # optional dict or pycolmap.RANSACOptions

Two-view geometry estimation

COLMAP can also estimate a relative pose between two calibrated cameras by estimating both E and H and accounting for the degeneracies of each model.

# Parameters:
# - camera1: pycolmap.Camera of image 1
# - points1: Nx2 array; 2D pixel coordinates in image 1
# - camera2: pycolmap.Camera of image 2
# - points2: Nx2 array; 2D pixel coordinates in image 2
# Optional parameters:
# - matches: Nx2 integer array; correspondences across images
# - options: dict or pycolmap.TwoViewGeometryOptions
answer = pycolmap.estimate_calibrated_two_view_geometry(camera1, points1, camera2, points2)
# Returns: pycolmap.TwoViewGeometry

The TwoViewGeometryOptions control how each model is selected. The output structure contains the geometric model, inlier matches, the relative pose (if options.compute_relative_pose=True), and the type of camera configuration, which is an instance of the enum pycolmap.TwoViewGeometryConfiguration.

Camera argument

Some estimators expect a COLMAP camera object, which can be created as follow:

camera = pycolmap.Camera(

The different camera models and their extra parameters are defined in colmap/src/colmap/sensor/models.h. For example for a pinhole camera:

camera = pycolmap.Camera(
    params=[focal_length, cx, cy],

Alternatively, we can also pass a camera dictionary:

camera_dict = {
    'width': IMAGE_WIDTH,
    'height': IMAGE_HEIGHT,

SIFT feature extraction

import numpy as np
import pycolmap
from PIL import Image, ImageOps

# Input should be grayscale image with range [0, 1].
img ='image.jpg').convert('RGB')
img = ImageOps.grayscale(img)
img = np.array(img).astype(np.float) / 255.

# Optional parameters:
# - options: dict or pycolmap.SiftExtractionOptions
# - device: default uses the GPU if available
sift = pycolmap.Sift()

# Parameters:
# - image: HxW float array
keypoints, descriptors = sift.extract(img)
# Returns:
# - keypoints: Nx4 array; format: x (j), y (i), scale, orientation
# - descriptors: Nx128 array; L2-normalized descriptors


  • Add documentation
  • Add more detailed examples
  • Add unit tests for reconstruction bindings

Created and maintained by Mihai Dusmanu, Philipp Lindenberger, John Lambert, Paul-Edouard Sarlin, and other contributors.

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distributions

No source distribution files available for this release.See tutorial on generating distribution archives.

Built Distributions

pycolmap-0.6.1-cp311-cp311-win_amd64.whl (13.0 MB view hashes)

Uploaded CPython 3.11 Windows x86-64

pycolmap-0.6.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (11.9 MB view hashes)

Uploaded CPython 3.11 manylinux: glibc 2.17+ x86-64

pycolmap-0.6.1-cp311-cp311-macosx_11_0_arm64.whl (7.9 MB view hashes)

Uploaded CPython 3.11 macOS 11.0+ ARM64

pycolmap-0.6.1-cp311-cp311-macosx_10_9_x86_64.whl (9.1 MB view hashes)

Uploaded CPython 3.11 macOS 10.9+ x86-64

pycolmap-0.6.1-cp310-cp310-win_amd64.whl (13.0 MB view hashes)

Uploaded CPython 3.10 Windows x86-64

pycolmap-0.6.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (11.9 MB view hashes)

Uploaded CPython 3.10 manylinux: glibc 2.17+ x86-64

pycolmap-0.6.1-cp310-cp310-macosx_11_0_arm64.whl (7.9 MB view hashes)

Uploaded CPython 3.10 macOS 11.0+ ARM64

pycolmap-0.6.1-cp310-cp310-macosx_10_9_x86_64.whl (9.1 MB view hashes)

Uploaded CPython 3.10 macOS 10.9+ x86-64

pycolmap-0.6.1-cp39-cp39-win_amd64.whl (13.0 MB view hashes)

Uploaded CPython 3.9 Windows x86-64

pycolmap-0.6.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (11.9 MB view hashes)

Uploaded CPython 3.9 manylinux: glibc 2.17+ x86-64

pycolmap-0.6.1-cp39-cp39-macosx_11_0_arm64.whl (7.9 MB view hashes)

Uploaded CPython 3.9 macOS 11.0+ ARM64

pycolmap-0.6.1-cp39-cp39-macosx_10_9_x86_64.whl (9.1 MB view hashes)

Uploaded CPython 3.9 macOS 10.9+ x86-64

pycolmap-0.6.1-cp38-cp38-win_amd64.whl (13.0 MB view hashes)

Uploaded CPython 3.8 Windows x86-64

pycolmap-0.6.1-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (11.9 MB view hashes)

Uploaded CPython 3.8 manylinux: glibc 2.17+ x86-64

pycolmap-0.6.1-cp38-cp38-macosx_11_0_arm64.whl (7.9 MB view hashes)

Uploaded CPython 3.8 macOS 11.0+ ARM64

pycolmap-0.6.1-cp38-cp38-macosx_10_9_x86_64.whl (9.1 MB view hashes)

Uploaded CPython 3.8 macOS 10.9+ x86-64

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page