V3D-SLAM: Robust RGB-D SLAM in Dynamic Environments with 3D Semantic Geometry Voting

TL;DR

V3D-SLAM introduces a robust RGB-D SLAM approach that effectively handles dynamic environments by identifying and removing moving objects through 3D geometry voting and noise refinement, outperforming recent methods.

Contribution

The paper presents a novel SLAM method that uses 3D semantic geometry voting and dynamic noise detection to improve robustness in dynamic scenes.

Findings

Outperforms recent state-of-the-art SLAM methods on TUM RGB-D benchmark.

Effectively identifies and removes moving objects in dynamic environments.

Enhances SLAM accuracy and robustness in highly dynamic scenes.

Abstract

Simultaneous localization and mapping (SLAM) in highly dynamic environments is challenging due to the correlation complexity between moving objects and the camera pose. Many methods have been proposed to deal with this problem; however, the moving properties of dynamic objects with a moving camera remain unclear. Therefore, to improve SLAM's performance, minimizing disruptive events of moving objects with a physical understanding of 3D shapes and dynamics of objects is needed. In this paper, we propose a robust method, V3D-SLAM, to remove moving objects via two lightweight re-evaluation stages, including identifying potentially moving and static objects using a spatial-reasoned Hough voting mechanism and refining static objects by detecting dynamic noise caused by intra-object motions using Chamfer distances as similarity measurements. Our experiment on the TUM RGB-D benchmark on dynamic sequences with ground-truth camera trajectories showed that our methods outperform the most recent state-of-the-art SLAM methods. Our source code is available at https://github.com/tuantdang/v3d-slam.