CAMAv2: A Vision-Centric Approach for Static Map Element Annotation

TL;DR

CAMAv2 is a vision-based framework that produces highly accurate, consistent 3D annotations of static map elements without using LiDAR, improving data quality for HD map construction and training.

Contribution

It introduces a novel vision-centric method for high-quality, spatial-temporally consistent map annotations without LiDAR, enhancing dataset accuracy and efficiency.

Findings

Lower reprojection errors compared to original annotations

Improved model training accuracy with CAMAv2 annotations

Achieved high spatial-temporal consistency across sequences

Abstract

The recent development of online static map element (a.k.a. HD map) construction algorithms has raised a vast demand for data with ground truth annotations. However, available public datasets currently cannot provide high-quality training data regarding consistency and accuracy. For instance, the manual labelled (low efficiency) nuScenes still contains misalignment and inconsistency between the HD maps and images (e.g., around 8.03 pixels reprojection error on average). To this end, we present CAMAv2: a vision-centric approach for Consistent and Accurate Map Annotation. Without LiDAR inputs, our proposed framework can still generate high-quality 3D annotations of static map elements. Specifically, the annotation can achieve high reprojection accuracy across all surrounding cameras and is spatial-temporal consistent across the whole sequence. We apply our proposed framework to the popular nuScenes dataset to provide efficient and highly accurate annotations. Compared with the original nuScenes static map element, our CAMAv2 annotations achieve lower reprojection errors (e.g., 4.96 vs. 8.03 pixels). Models trained with annotations from CAMAv2 also achieve lower reprojection errors (e.g., 5.62 vs. 8.43 pixels).