APRIL-ZJU/Gaussian-LIC

[ICRA 2025 & IJRR 2026] Gaussian-LIC2: LiDAR-Inertial-Camera Gaussian Splatting SLAM (in Real Time)

What it solves

Gaussian-LIC2 addresses the challenge of creating high-fidelity, photo-realistic 3D maps in real time while maintaining accurate pose estimation for robots. It combines the strengths of LiDAR, inertial sensors, and cameras to overcome the limitations of individual sensors in SLAM (Simultaneous Localization and Mapping) systems.

How it works

The system integrates LiDAR-Inertial-Camera fusion with Gaussian Splatting. It simultaneously performs robust pose estimation and constructs a 3D Gaussian map. The pipeline leverages TensorRT for deployment and utilizes a combination of third-party libraries like OpenCV and LibTorch to process sensor data and render the environment in real time.

Who it’s for

This project is designed for robotics researchers and engineers working on SLAM, 3D reconstruction, and autonomous navigation who require photo-realistic environment mapping and precise localization.

Highlights

  • Multi-Sensor Fusion: Integrates LiDAR, Inertial, and Camera data for increased robustness.
  • Photo-Realistic Mapping: Uses Gaussian Splatting to create high-quality 3D maps.
  • Real-Time Performance: Capable of simultaneous pose estimation and map construction in real time.
  • Academic Validation: Accepted by ICRA 2025 and the International Journal of Robotics Research (IJRR) 2026.

Related

  • Project
    ouguangjun/Leg-KILO

    A real-time LiDAR-based SLAM system designed for high-dynamic platforms like legged robots, featuring a two-stage ESKF frontend and factor graph backend for robust mapping.

  • Project
    hku-mars/GS-SDF

    A LiDAR-augmented system that combines Gaussian Splatting and neural Signed Distance Fields (SDF) for geometrically consistent photorealistic rendering and 3D surface reconstruction.

  • Project
    PRBonn/kiss-slam

    A simple and robust 3D LiDAR SLAM system designed for accurate simultaneous localization and mapping with high generalization capabilities.

  • Project
    SJTU-ViSYS/Ground-Fusion

    A low-cost sensor-fusion SLAM system for ground vehicles that integrates RGB-D, IMU, wheel odometry, and GNSS to maintain robust localization during sensor failures.

  • Project
    rpng/MINS

    A multisensor-aided inertial navigation system that flexibly fuses IMU, wheel encoders, cameras, GNSS, and LiDAR for robust robot localization and online sensor calibration.