InsMapper: Exploring Inner-instance Information for Vectorized HD Mapping

TL;DR

InsMapper introduces a transformer-based approach that leverages inner-instance information to improve vectorized HD map detection, significantly outperforming previous methods on challenging datasets.

Contribution

The paper proposes InsMapper, a novel system utilizing inner-instance information with three key designs to enhance vectorized HD map detection using transformers.

Findings

Outperforms previous state-of-the-art on NuScenes and Argoverse 2 datasets.

Effectively harnesses inner-instance information for better line detection.

Demonstrates high adaptability to recent HD map detection frameworks.

Abstract

Vectorized high-definition (HD) maps contain detailed information about surrounding road elements, which are crucial for various downstream tasks in modern autonomous vehicles, such as motion planning and vehicle control. Recent works attempt to directly detect the vectorized HD map as a point set prediction task, achieving notable detection performance improvements. However, these methods usually overlook and fail to analyze the important inner-instance correlations between predicted points, impeding further advancements. To address this issue, we investigate the utilization of inner-instance information for vectorized high-definition mapping through transformers, and propose a powerful system named $\textbf{InsMapper}$, which effectively harnesses inner-instance information with three exquisite designs, including hybrid query generation, inner-instance query fusion, and inner-instance feature aggregation. The first two modules can better initialize queries for line detection, while the last one refines predicted line instances. InsMapper is highly adaptable and can be seamlessly modified to align with the most recent HD map detection frameworks. Extensive experimental evaluations are conducted on the challenging NuScenes and Argoverse 2 datasets, where InsMapper surpasses the previous state-of-the-art method, demonstrating its effectiveness and generality. The project page for this work is available at https://tonyxuqaq.github.io/InsMapper/ .