Gaussian-LIC2: LiDAR-Inertial-Camera
Gaussian Splatting SLAM

ICRA 2025 & IJRR 2026

Gaussian-LIC2 is a photo-realistic LiDAR-Inertial-Camera Gaussian Splatting SLAM system, which simultaneously performs robust, accurate pose estimation and constructs a photo-realistic 3D Gaussian map in real time.

📢 News

• [2026-07-25] Gaussian-LIC2 is accepted by the International Journal of Robotics Research (IJRR)! 🎉
• [2026-02-21] Gaussian-LIC2 is released! 🚀 (stay tuned for updates)
• [2025-07-08] Gaussian-LIC2 is unveiled! 🎉 [Paper] [Page] [YouTube] [bilibili]
• [2025-07-07] The enhanced version of the Gaussian-LIC code is released!
• [2025-01-28] Gaussian-LIC is accepted to ICRA 2025! 🎉
• [2024-09-26] The second version of the paper is available on arXiv.
• [2024-04-10] The first version of the paper is available on arXiv.

💌 Contact

Questions? Please don't hesitate to reach out to Xiaolei Lang (Jerry) at jerry_locker@zju.edu.cn.

Install

We test on ubuntu 20.04 with an NVIDIA RTX 3090 / 4090.

1. Exit Conda environment.

2. Prepare third-party libraries according to Coco-LIC.

3. Install CUDA 11.7 with cuDNN v8.9.7.

4. Build OpenCV 4.7.0.（must be built with opencv_contrib 4.7.0 and CUDA, no installation required）

   mkdir -p ~/Software/opencv
   cd ~/Software/opencv
   wget https://github.com/opencv/opencv/archive/refs/tags/4.7.0.tar.gz && tar -zxvf 4.7.0.tar.gz && rm -rf 4.7.0.tar.gz
   wget https://github.com/opencv/opencv_contrib/archive/refs/tags/4.7.0.tar.gz && tar -zxvf 4.7.0.tar.gz && rm -rf 4.7.0.tar.gz
   
   cd ~/Software/opencv/opencv-4.7.0
   mkdir build && cd build
   cmake -DCMAKE_BUILD_TYPE=RELEASE -DWITH_CUDA=ON -DWITH_CUDNN=ON -DOPENCV_DNN_CUDA=ON -DWITH_NVCUVID=ON -DCUDA_TOOLKIT_ROOT_DIR=/usr/local/cuda-11.7 -DOPENCV_EXTRA_MODULES_PATH="../../opencv_contrib-4.7.0/modules" -DBUILD_TIFF=ON -DBUILD_ZLIB=ON -DBUILD_JASPER=ON -DBUILD_CCALIB=ON -DBUILD_JPEG=ON -DWITH_FFMPEG=ON ..
   make -j$(nproc)

5. Prepare LibTorch.（no compilation or installation required）

   cd ~/Software
   wget https://download.pytorch.org/libtorch/cu117/libtorch-cxx11-abi-shared-with-deps-2.0.1%2Bcu117.zip
   unzip libtorch-cxx11-abi-shared-with-deps-2.0.1+cu117.zip && rm -rf libtorch-cxx11-abi-shared-with-deps-2.0.1+cu117.zip

6. Prepare TensorRT.（no compilation or installation required）

   cd ~/Software
   wget https://developer.nvidia.com/downloads/compute/machine-learning/tensorrt/secure/8.6.1/tars/TensorRT-8.6.1.6.Linux.x86_64-gnu.cuda-11.8.tar.gz
   tar -zxvf TensorRT-8.6.1.6.Linux.x86_64-gnu.cuda-11.8.tar.gz && rm -rf TensorRT-8.6.1.6.Linux.x86_64-gnu.cuda-11.8.tar.gz

7. Install Coco-LIC.

   mkdir -p ~/catkin_coco/src
   cd ~/catkin_coco/src
   git clone https://github.com/Livox-SDK/livox_ros_driver.git
   cd ~/catkin_coco && catkin_make
   cd ~/catkin_coco/src
   git clone https://github.com/APRIL-ZJU/Coco-LIC.git
   cd ~/catkin_coco && catkin_make

8. Install Gaussian-LIC.

   mkdir -p ~/catkin_gaussian/src
   cd ~/catkin_gaussian/src
   git clone https://github.com/APRIL-ZJU/Gaussian-LIC.git
   cd ~/catkin_gaussian && catkin_make

9. TensorRT Deployment.

   download and save Large_300.pth to ~/catkin_gaussian/src/Gaussian-LIC/ckpt.

   cd ~/catkin_gaussian/src/Gaussian-LIC/ckpt
   
   chmod +x setup_spnet.sh
   ./setup_spnet.sh
   
   chmod +x export_onnx.sh
   ./export_onnx.sh
   
   chmod +x build_trt.sh
   ./build_trt.sh

Run

Quick start on the sequence CBD_Building_01 in the FAST-LIVO2 dataset.

• Download FAST-LIVO Dataset or FAST-LIVO2 Dataset or R3LIVE Dataset or MCD Dataset or M2DGR Dataset.

• Modify bag_path in the config/ct_odometry_fastlivo2.yaml file of Coco-LIC.

• Launch Gaussian-LIC.

  cd ~/catkin_gaussian
  source devel/setup.bash
  roslaunch gaussian_lic fastlivo2.launch  // The terminal will print "😋 Gaussian-LIC Ready!".

• Launch Coco-LIC.

  Note：For real-time use and runtime analysis, please turn off the rviz in Coco-LIC by commenting the sentence <node pkg="rviz" type="rviz" name="rviz_odom" output="log" required = "true" args="-d $(find cocolic)/config/coco.rviz" /> in odometry.launch.

  cd ~/catkin_coco
  source devel/setup.bash
  roslaunch cocolic odometry.launch config_path:=config/ct_odometry_fastlivo2.yaml

• The mapping and rendering results will be saved in ~/catkin_gaussian/src/Gaussian-LIC/result.

Checklist

• Support fast post-optimization
• Release the optimized Coco-LIC
• Provide the dockerfile
• Release the meshing tools
• Release our Gaussian-LIC2 dataset

Citation

If you find our work helpful, please consider citing 🌟:

@inproceedings{lang2025gaussian,
  title={Gaussian-LIC: Real-time photo-realistic SLAM with Gaussian splatting and LiDAR-inertial-camera fusion},
  author={Lang, Xiaolei and Li, Laijian and Wu, Chenming and Zhao, Chen and Liu, Lina and Liu, Yong and Lv, Jiajun and Zuo, Xingxing},
  booktitle={2025 IEEE International Conference on Robotics and Automation (ICRA)},
  pages={8500--8507},
  year={2025},
  organization={IEEE}
}

@article{lang2026gaussian2,
  title={Gaussian-LIC2: LiDAR-Inertial-Camera Gaussian Splatting SLAM}, 
  author={Lang, Xiaolei and Lv, Jiajun and Tang, Kai and Li, Laijian and Huang, Jianxin and Liu, Lina and Liu, Yong and Zuo, Xingxing},
  journal={The International Journal of Robotics Research},
  year={2026}
}

Acknowledgement

Thanks for 3DGS, Taming-3DGS, StopThePop, Photo-SLAM and SPNet.

LICENSE

The code is released under the GNU General Public License v3 (GPL-3).