MapTracker: Tracking with Strided Memory Fusion for Consistent Vector HD Mapping

TL;DR

MapTracker introduces a novel vector HD-mapping algorithm that formulates mapping as a tracking problem, leveraging memory buffers of raster and vector latents to ensure consistent, temporally aligned reconstructions of road elements over time.

Contribution

The paper proposes MapTracker, a new method that fuses strided memory latents for improved temporal consistency in HD-mapping, with benchmark improvements on nuScenes and Agroverse2 datasets.

Findings

Outperforms existing methods by over 8% on nuScenes

Achieves over 19% improvement on Agroverse2 with consistency metrics

Provides enhanced processing code and augmented metrics for better benchmarking

Abstract

This paper presents a vector HD-mapping algorithm that formulates the mapping as a tracking task and uses a history of memory latents to ensure consistent reconstructions over time. Our method, MapTracker, accumulates a sensor stream into memory buffers of two latent representations: 1) Raster latents in the bird's-eye-view (BEV) space and 2) Vector latents over the road elements (i.e., pedestrian-crossings, lane-dividers, and road-boundaries). The approach borrows the query propagation paradigm from the tracking literature that explicitly associates tracked road elements from the previous frame to the current, while fusing a subset of memory latents selected with distance strides to further enhance temporal consistency. A vector latent is decoded to reconstruct the geometry of a road element. The paper further makes benchmark contributions by 1) Improving processing code for existing datasets to produce consistent ground truth with temporal alignments and 2) Augmenting existing mAP metrics with consistency checks. MapTracker significantly outperforms existing methods on both nuScenes and Agroverse2 datasets by over 8% and 19% on the conventional and the new consistency-aware metrics, respectively. The code and models are available on our project page: https://map-tracker.github.io.