MapSeg: Segmentation guided structured model for online HD map construction
This work addresses the need for real-time, accurate HD maps in autonomous driving and intelligent transportation, but it is incremental as it builds upon existing methods like MapTR.
The paper tackles the problem of online high-definition map construction by proposing MapSeg, a segmentation-guided structured model that improves accuracy by incorporating global semantic information, resulting in enhanced performance for applications like autonomous driving.
The development of online high-definition maps is significant since they provide real-time, accurate, and updatable geographic information for location-based applications, such as autonomous driving and intelligent transportation, thus improving the performance and reliability of these applications. Previous works, such as VectorMapNet and MapTR, show that direct model generation of vectorized HD maps is a promising solution. However, these methods did not take into account the usage of global semantic information to improve map construction accuracy. To address this limitation, we propose a segmentation-guided structured model (MapSeg) for online HD map construction. Specifically, we added a UV segmentation module (USM) and a BEV segmentation module (BSM) based on the MapTR structure, enabling the model to better capture the semantic information.