通过双向对齐提升高精地图时序一致性，减少动态环境下的地图抖动。
MapTCL: Temporal Consistency Learning via Bidirectional Alignment for Vectorized HD Map Construction
- 用双向向量对齐建模前后帧间几何语义差异，作为辅助损失。
- 在nuScenes和Argoverse 2上分别提升3.7/2.8和3.1/2.5 mAP/C-mAP。
- 无需额外推理开销，可直接嵌入现有模型提升稳定性。
在动态城市环境中构建可靠的在线高精地图仍具挑战，主要源于移动物体与遮挡。现有方法虽采用特征级时序融合，但仅依赖每帧的真值监督，缺乏显式目标来惩罚连续地图间的几何噪声与时间抖动。为此，我们提出MapTCL，一种基于双向对齐的时序一致性学习策略。具体地，双向向量一致性学习（BVCL）将关联的过去与当前向量实例间的几何与语义差异建模为辅助损失；同时引入栅格地图一致性学习（RCL）稳定密集BEV特征。联合训练双损失后，显著提升了生成地图的时序稳定性。在两个标准基准上大量实验表明，MapTCL作为通用插件模块，能持续提升基线模型性能，在nuScenes上实现+3.7 mAP & +2.8 C-mAP，Argoverse 2上实现+3.1 mAP & +2.5 C-mAP，且无额外推理开销。
原文摘要 · Abstract (English)
Constructing reliable online HD maps remains challenging in dynamic urban environments due to moving objects and occlusions. While recent works employ feature-level temporal fusion to address this, they rely solely on per-frame ground truth supervision. Consequently, they lack an explicit objective to directly penalize the geometric noise and temporal jitter between consecutive online HD maps. To address this, we propose MapTCL, an auxiliary training strategy that formulates temporal consistency loss between current and past frames via bidirectional alignment. Specifically, Bidirectional Vector Consistency Learning (BVCL) models the geometric and semantic discrepancies between associated past and current vector instances as an auxiliary loss. We also employ Raster map Consistency Learning (RCL) as an additional loss to stabilize dense BEV features. By jointly training with these dual losses, MapTCL improves the temporal stability of generated HD maps. Extensive experiments on two standard benchmarks demonstrate the effectiveness of our approach. As a versatile plug-and-play module, MapTCL consistently enhances existing baseline models, achieving gains of +3.7 mAP & +2.8 C-mAP on nuScenes and +3.1 mAP & +2.5 C-mAP on Argoverse 2 without additional inference overhead.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。