Model based RGBD SLAM

Kathia Melbouci (1) , Sylvie Naudet Collette (1) , Vincent Gay-Bellile (1) , Omar Ait-Aider (2) , Michel Dhome (3)

1 LVIC - Laboratoire Vision et Ingénierie des Contenus
2 IP - Institut Pascal
3 IP - Institut Pascal

In this paper we propose to improve the localization and the 3D mapping provided by an RGBD SLAM algorithm, using a prior knowledge of the 3D model of the environment. The proposed solution relies on an feature-based RGBD SLAM algorithm to localize the camera and update the 3D map of the scene. To improve the accuracy and the robustness of the localization, we propose to combine in a local bundle adjustment process, geometric information provided by a prior coarse 3D model of the scene (e.g. generated from the 2D plan of the building) with RGBD data. The proposed approach is evaluated on a public benchmark dataset as well as on a real scene acquired by a Kinect sensor.

Keywords: RGBD slam; 3D model; Local bundle adjustment

Domains: Engineering Sciences [physics]; Signal and Image processing

Kathia Melbouci, Sylvie Naudet Collette, Vincent Gay-Bellile, Omar Ait-Aider, Michel Dhome. Model based RGBD SLAM. 2016 IEEE International Conference on Image Processing (ICIP), Sep 2016, Phoenix, United States. ⟨10.1109/ICIP.2016.7532833⟩. ⟨cea-01830443⟩