We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations

DBLP - CS Bibliography


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Computer Vision and Pattern Recognition

Title: MASS: Multi-Attentional Semantic Segmentation of LiDAR Data for Dense Top-View Understanding

Abstract: At the heart of all automated driving systems is the ability to sense the surroundings, e.g., through semantic segmentation of LiDAR sequences, which experienced a remarkable progress due to the release of large datasets such as SemanticKITTI and nuScenes-LidarSeg. While most previous works focus on sparse segmentation of the LiDAR input, dense output masks provide self-driving cars with almost complete environment information. In this paper, we introduce MASS - a Multi-Attentional Semantic Segmentation model specifically built for dense top-view understanding of the driving scenes. Our framework operates on pillar- and occupancy features and comprises three attention-based building blocks: (1) a keypoint-driven graph attention, (2) an LSTM-based attention computed from a vector embedding of the spatial input, and (3) a pillar-based attention, resulting in a dense 360-degree segmentation mask. With extensive experiments on both, SemanticKITTI and nuScenes-LidarSeg, we quantitatively demonstrate the effectiveness of our model, outperforming the state of the art by 19.0% on SemanticKITTI and reaching 30.4% in mIoU on nuScenes-LidarSeg, where MASS is the first work addressing the dense segmentation task. Furthermore, our multi-attention model is shown to be very effective for 3D object detection validated on the KITTI-3D dataset, showcasing its high generalizability to other tasks related to 3D vision.
Comments: Accepted to IEEE Transactions on Intelligent Transportation Systems (T-ITS). Code is publicly available at this https URL
Subjects: Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO)
Cite as: arXiv:2107.00346 [cs.CV]
  (or arXiv:2107.00346v2 [cs.CV] for this version)

Submission history

From: Kailun Yang [view email]
[v1] Thu, 1 Jul 2021 10:19:32 GMT (2958kb,D)
[v2] Thu, 20 Jan 2022 15:40:28 GMT (5788kb,D)

Link back to: arXiv, form interface, contact.