4D Panoptic LiDAR Segmentation

Mehmet Aygun; Aljosa Osep; Mark Weber; Maxim Maximov; Cyrill Stachniss; Jens Behley; Laura Leal-Taixé

doi:10.1109/cvpr46437.2021.00548

CVPR2021顶会

4D Panoptic LiDAR Segmentation

Mehmet Aygun, Aljosa Osep, Mark Weber, Maxim Maximov, Cyrill Stachniss, Jens Behley, Laura Leal-Taixé

DOI 出版方

2021年份

18顶会引用

摘要

instances semantics semantics semantic segmentation panoptic segmentation multi-object tracking 4D panoptic segmentation Figure 1: Types of LiDAR-based scene understanding. Semantic and panoptic segmentation assign semantic classes and determine instances in 3D space. Multi-object tracking encompasses 3D object detection in space, followed by association over time. 4D panoptic LiDAR segmentation jointly tackles semantic and instance segmentation in 3D space over time.