stella-cv/stella_vslam

Health v2 · maintenance only

Flags: no_releases no_license

Adoption not part of the score

1238 stars · 468 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

stella_vslam is a community-maintained fork of OpenVSLAM implementing a monocular, stereo, and RGBD visual SLAM system in C++. It supports multiple camera models (perspective, fisheye, equirectangular), can store and reload maps for relocalization, and offers a modular API with a separate ROS wrapper.

Use cases

• estimate camera pose from video with visual SLAM
• build a 3D map from monocular or stereo camera footage
• localize new images against a prebuilt SLAM map
• run SLAM with a 360-degree equirectangular camera
• add visual odometry to a robot with ROS
• track camera position for AR or drone navigation

When to choose

• you need a modular visual SLAM library supporting many camera models
• you want to store, load, and relocalize against prebuilt maps
• you use fisheye or equirectangular cameras like RICOH THETA or insta360
• you want an actively maintained OpenVSLAM successor with a ROS wrapper

When to avoid

• you need a permissively licensed codebase without reservation - check the license history first
• you require dense reconstruction or depth maps rather than sparse feature-based SLAM
• you need a turnkey application rather than a C++ library to integrate
• you depend on the original OpenVSLAM repository, which is no longer available

Facets

library · maturity active

computer-vision simulation sdk computer-vision robotics autonomous-vehicles cpp cross-platform visual-slam slam monocular-camera stereo-camera rgbd-camera camera-localization mapping openvslam-fork equirectangular fisheye algorithms linux macos

Member repositories

For agents

markdown · JSON · MCP: product_card(name="stella-cv/stella_vslam")