opengv

A collection of computer vision methods for geometric vision problems. It covers camera pose estimation, triangulation and point cloud alignment for calibrated cameras.

Share on XCustom license (see repository)

Overview

OpenGV is a collection of computer vision methods for geometric vision problems. It covers absolute pose, relative pose, triangulation and point cloud alignment for calibrated cameras. Every problem can be solved for central or non-central cameras and used inside a random sample consensus or nonlinear optimization setup. Matlab and Python interfaces are also provided.

Key features

  • Absolute pose and relative pose estimation
  • Triangulation and point cloud alignment
  • Supports central and non-central cameras
  • Matlab and Python interfaces

Best for

Computer vision researchers and engineers who need geometric solvers for calibrated cameras, including multi-camera rigs. The methods target the calibrated case.

Upstream
mapillary/opengv
Fork on GitHub
Guo-astro/opengv
Upstream stars
5
Category
Finance, data and research
License
Custom license (see repository)
Forked
2020-12-06
Sync status
In syncLast synced 2026-09-29