In autonomous driving, the high-definition (HD) map plays a crucial role in localization and planning. Recently, several methods have facilitated end-to-end online map construction in DETR-like frameworks. However, little attention has been paid to the potential capabilities of exploring the query mechanism. This paper introduces MapQR, an end-to-end method with an emphasis on enhancing query capabilities for constructing online vectorized maps. Although the map construction is essentially a point set prediction task, MapQR utilizes instance queries rather than point queries. These instance queries are scattered for the prediction of point sets and subsequently gathered for the final matching. This query design, called the scatter-and-gather query, shares content information in the same map element and avoids possible inconsistency of content information in point queries. We further exploit prior information to enhance an instance query by adding positional information embedded from their reference points. Together with a simple and effective improvement of a BEV encoder, the proposed MapQR achieves the best mean average precision (mAP) and maintains good efficiency on both nuScenes and Argoverse 2. In addition, integrating our query design into other models can boost their performance significantly. The code will be available at https://github.com/HXMap/MapQR.

翻译：在自动驾驶中，高清地图在定位与规划中扮演着关键角色。近期，多种方法已在类似DETR的框架中实现了端到端的在线地图构建，但查询机制的潜在能力尚未得到充分探索。本文提出MapQR——一种聚焦于增强查询能力的端到端矢量化地图构建方法。尽管地图构建本质上是点集预测任务，MapQR采用实例查询而非点查询：这些实例查询被分散用于点集预测，随后再聚合以完成最终匹配。这种"分散-聚合"查询设计使得同一地图元素内的内容信息得以共享，避免了点查询可能出现的内容信息不一致问题。我们进一步利用先验信息，通过嵌入参考点的位置信息来增强实例查询。结合简洁高效的BEV编码器改进，所提出的MapQR在nuScenes和Argoverse 2数据集上均取得了最佳平均精度（mAP）并保持了良好的效率。此外，将该查询设计集成至其他模型可显著提升其性能。代码将发布于https://github.com/HXMap/MapQR。