This is the implementation of our paper: "Probabilistic Dense Reconstruction from a Moving Camera" in IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2018. Our implementation runs on Linux and is fully integrated with ROS (developed in version of indigo). Eigen, OpenCV 3+, and CUDA are required.

The high resolution video of our submitted paper is: https://1drv.ms/v/s!ApzRxvwAxXqQmlW9ZOrp9hdA7ude (720p is recommmanded)

To complie this code, you should modify the arch code (i.e. --generate-code arch=compute_XX,code=sm_XX) in the CMakeList.txt such that it adapts to your GPU.

The source code is released under GPLv3 license. If you use our code, please cite our paper:

[1] Yonggen Ling, Kaixuan Wang and Shaojie Shen, "Probabilistic Dense Reconstruction from a Moving Camera" in IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2018.

For more questions, please contact ylingaa at connect dot ust dot hk .