hku-mars/r3live

Health v2 · maintenance only

Adoption not part of the score

2442 stars · 486 forks observed · 2026-08-28

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

R3LIVE is a tightly-coupled LiDAR-Inertial-Visual sensor fusion framework for robust, real-time state estimation and RGB-colored 3D mapping. It combines a LiDAR-inertial odometry subsystem (FAST-LIO) that builds geometric structure with a visual-inertial odometry subsystem that renders color texture onto the global map.

Use cases

• build rgb-colored 3d maps from lidar and camera data
• real-time lidar-inertial-visual odometry for a handheld scanner
• state estimation in lidar-degenerated environments
• reconstruct colored meshes of indoor and outdoor scenes
• fuse lidar imu and camera measurements for slam
• generate dense 3d point cloud maps with color texture

When to choose

• you need tightly-coupled fusion of lidar, imu, and camera for accurate odometry
• you want rgb-colored 3d maps rather than bare geometric point clouds
• your scenario includes lidar-degenerated environments like tunnels or corridors
• you are doing academic research on slam and sensor fusion

When to avoid

• you only have a camera without lidar or imu sensors
• you need a commercially-licensed solution since the code is GPLv2 with restrictions on commercial use
• you need a lightweight odometry-only solution without mapping
• your platform is not linux/ros based

Facets

library · maturity active

computer-vision image-processing graphics simulation sdk robotics computer-vision autonomous-vehicles cpp slam lidar-inertial-odometry visual-inertial-odometry sensor-fusion rgb-colored-mapping 3d-reconstruction mesh-reconstruction lidar-camera-fusion state-estimation robotics algorithms linux ros

Member repositories

For agents

markdown · JSON · MCP: product_card(name="hku-mars/r3live")