LidarMultiNet: Unifying LiDAR Semantic Segmentation, 3D Object Detection, and Panoptic Segmentation in a Single Multi-task Network
This work addresses the need for efficient and integrated perception systems in autonomous driving by combining multiple tasks into one network, though it is incremental as it builds on existing voxel-based methods.
The authors tackled the problem of unifying multiple LiDAR perception tasks (semantic segmentation, object detection, panoptic segmentation) into a single network, achieving a mIoU of 71.13 and winning the Waymo Open Dataset 3D semantic segmentation challenge 2022.
This technical report presents the 1st place winning solution for the Waymo Open Dataset 3D semantic segmentation challenge 2022. Our network, termed LidarMultiNet, unifies the major LiDAR perception tasks such as 3D semantic segmentation, object detection, and panoptic segmentation in a single framework. At the core of LidarMultiNet is a strong 3D voxel-based encoder-decoder network with a novel Global Context Pooling (GCP) module extracting global contextual features from a LiDAR frame to complement its local features. An optional second stage is proposed to refine the first-stage segmentation or generate accurate panoptic segmentation results. Our solution achieves a mIoU of 71.13 and is the best for most of the 22 classes on the Waymo 3D semantic segmentation test set, outperforming all the other 3D semantic segmentation methods on the official leaderboard. We demonstrate for the first time that major LiDAR perception tasks can be unified in a single strong network that can be trained end-to-end.