RainerKuemmerle/g2o

Health v2 · maintenance only

Flags: no_license

Adoption not part of the score

3461 stars · 1157 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

g2o is an open-source C++ framework for optimizing graph-based nonlinear error functions, commonly used for nonlinear least squares problems. It is designed to be easily extensible and provides solutions for variants of SLAM (simultaneous localization and mapping) and bundle adjustment (BA).

Use cases

• solve simultaneous localization and mapping (SLAM) problems
• perform bundle adjustment for structure-from-motion
• optimize graph-based nonlinear least squares problems
• estimate robot trajectories from sensor measurements
• build custom state estimation solvers in C++
• minimize error functions over pose graphs

When to choose

• you need a fast, extensible C++ solver for graph-based nonlinear optimization
• you are working on SLAM, visual odometry, or bundle adjustment in robotics or computer vision
• you want a well-established library with performance comparable to state-of-the-art approaches

When to avoid

• you need a Python-first optimization library (Python bindings are experimental on a separate branch)
• your problem is not expressible as a sparse graph of nonlinear error terms
• you require a permissively licensed dependency and license terms are critical (no license is listed in the repository)

Facets

library · maturity stable

math simulation computer-vision robotics robotics computer-vision autonomous-vehicles windows cpp cross-platform graph-optimization slam bundle-adjustment nonlinear-least-squares sparse-solver c-plus-plus algorithms linux macos

Member repositories

For agents

markdown · JSON · MCP: product_card(name="RainerKuemmerle/g2o")