This code is for paper "Fusing Semantic Segmentation and Object Detection for Visual SLAM in Dynamic Scenes".
Our system runs in the same environment as DS-SLAM, so it is necessary to configure environments according to the steps described in the DS-SLAM. If you encounter any difficulties in configuring the operating environment of DS-SLAM, please ask us questions. We also provide a DS-SLAM installation document in Chinese(clike me)[https://blog.csdn.net/qq_34131212/article/details/106803808?spm=1001.2014.3001.5501].
We used YOLO v4 in the experiment. See YOLO v4 for a detailed configuration process.