AerialFusionMapNet: Online HD Map Construction with Aerial-Onboard BEV Fusion
For autonomous driving perception, this work shows that a structured training design can better exploit aerial imagery for online HD map construction, outperforming more complex fusion architectures.
AerialFusionMapNet introduces a structured two-stage training strategy for fusing aerial and onboard sensor data to construct HD maps online, achieving 54.7 mAP on nuScenes, a 12.1% relative improvement over prior aerial-onboard fusion baselines.
High-resolution aerial imagery has recently emerged as a complementary modality for automated driving perception and has shown potential to improve birds-eye-view (BEV) scene understanding when fused with onboard sensors. Prior work demonstrated performance gains for online high-definition (HD) map construction through aerial-onboard fusion; however, conventional end-to-end fusion does not fully exploit the structural information contained in aerial representations. In this work, we introduce AerialFusionMapNet, a fusion-based mapping framework with a structured two-stage training strategy that explicitly enhances the contribution of aerial features within a unified pipeline. The proposed training scheme enables more effective integration of structural aerial priors. On the nuScenes geographic split, AerialFusionMapNet achieves up to 54.7 mAP, improving over prior aerial-onboard fusion baselines from 48.8 mAP by +5.9 absolute and +12.1% relative. The results suggest that structured training design, rather than increased architectural complexity, plays a more decisive role in unlocking the full potential of aerial imagery for online HD map construction. Code and trained models are available at https://github.com/DriverlessMobility/AerialFusionMapNet.