HD Maps are Lane Detection Generalizers: A Novel Generative Framework for Single-Source Domain Generalization
This work addresses the domain generalization challenge for lane detection in autonomous vehicles, offering an incremental improvement over existing methods.
The paper tackles the problem of poor generalization in lane detection models across different road environments by proposing a generative framework that uses HD Maps to create diverse training images, resulting in a 3.01% accuracy improvement over a domain adaptation model without accessing target domain data.
Lane detection is a vital task for vehicles to navigate and localize their position on the road. To ensure reliable driving, lane detection models must have robust generalization performance in various road environments. However, despite the advanced performance in the trained domain, their generalization performance still falls short of expectations due to the domain discrepancy. To bridge this gap, we propose a novel generative framework using HD Maps for Single-Source Domain Generalization (SSDG) in lane detection. We first generate numerous front-view images from lane markings of HD Maps. Next, we strategically select a core subset among the generated images using (i) lane structure and (ii) road surrounding criteria to maximize their diversity. In the end, utilizing this core set, we train lane detection models to boost their generalization performance. We validate that our generative framework from HD Maps outperforms the Domain Adaptation model MLDA with +3.01%p accuracy improvement, even though we do not access the target domain images.