TL;DR
Mask2Map is an innovative end-to-end method for online high-definition map construction in autonomous driving, utilizing bird's eye view segmentation masks and novel training techniques to improve accuracy over previous methods.
Contribution
The paper introduces Mask2Map, a new approach combining instance-level mask prediction and mask-driven map prediction, along with a denoising training method to enhance inter-network consistency.
Findings
Achieves 10.1% mAP improvement on nuScenes
Achieves 4.1 mAP improvement on Argoverse2
Demonstrates superior performance over state-of-the-art methods
Abstract
In this paper, we introduce Mask2Map, a novel end-to-end online HD map construction method designed for autonomous driving applications. Our approach focuses on predicting the class and ordered point set of map instances within a scene, represented in the bird's eye view (BEV). Mask2Map consists of two primary components: the Instance-Level Mask Prediction Network (IMPNet) and the Mask-Driven Map Prediction Network (MMPNet). IMPNet generates Mask-Aware Queries and BEV Segmentation Masks to capture comprehensive semantic information globally. Subsequently, MMPNet enhances these query features using local contextual information through two submodules: the Positional Query Generator (PQG) and the Geometric Feature Extractor (GFE). PQG extracts instance-level positional queries by embedding BEV positional information into Mask-Aware Queries, while GFE utilizes BEV Segmentation Masks to…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
MethodsSparse Evolutionary Training
