Semantic-Segmantation-based-Dynamic-Robust-SLAM

Semantic-Segmantation-based-Dynamic-Robust-SLAM

1. Compiling the ORB_SLAM2 follow the README.md, and start a RGBD node.
2. Putting the semantic_slam in ROS workspace, then use roslaunch to launch the semantic segmentation node. Download the model in model trained on ade20k / model trained on sunrgbd / and put them in models.
3. Running a .bag file in TUM database to publish rgb and depth images.

Acknowledgement

This work cannot be done without many open source projets. Special thanks to
semantic_slam
ORB_SLAM2
ORB_SLAM2_SSD_Semantic

License

This project is released under a GPLv3 license.