Paper presented at WACV 2023

As a great start to this year, TK researchers presented their work on “Unsupervised 4D LiDAR Moving Object Segmentation in Stationary Settings with Multivariate Occupancy Time Series” at this year's IEEE/CVF

Winter Conference on Applications of Computer Vision (WACV 2023). The paper proposes an unsupervised machine-learning approach for LiDAR moving object segmentation (MOS) focused especially on stationary LiDAR scenes. LiDAR MOS is of practical use in smart cities where LiDAR sensors can be mounted on, for instance, street lamps that cover a large city area [1], and moving objects have to be identified.

Citation info:

Thomas Kreutz, Max Mühlhäuser, and Alejandro Sanchez Guinea. 2023. Unsupervised 4D LiDAR Moving Object Segmentation in Stationary Settings with Multivariate Occupancy Time Series. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), IEEE, 1644–1653. CVF OpenAccess: https://openaccess.thecvf.com/content/WACV2023/html/Kreutz_Unsupervised_4D_LiDAR_Moving_Object_Segmentation_in_Stationary_Settings_With_WACV_2023_paper.html

[1] Mühlhäuser et al., Street Lamps as a Platform, Communications of the ACM, 2020