PrevPredMap: Exploring Temporal Modeling with Previous Predictions for Online Vectorized HD Map Construction

TL;DR

PrevPredMap is a novel framework that uses previous predictions for improved online vectorized HD map construction, excelling in long-term temporal modeling and map prior integration.

Contribution

It introduces PrevPredMap, a new temporal modeling approach leveraging previous predictions with specialized modules for enhanced HD map construction.

Findings

Achieves state-of-the-art results on nuScenes dataset

Demonstrates robustness in both single-frame and temporal modes

Outperforms existing methods in online vectorized HD map tasks

Abstract

Temporal information is crucial for detecting occluded instances. Existing temporal representations have progressed from BEV or PV features to more compact query features. Compared to these aforementioned features, predictions offer the highest level of abstraction, providing explicit information. In the context of online vectorized HD map construction, this unique characteristic of predictions is potentially advantageous for long-term temporal modeling and the integration of map priors. This paper introduces PrevPredMap, a pioneering temporal modeling framework that leverages previous predictions for constructing online vectorized HD maps. We have meticulously crafted two essential modules for PrevPredMap: the previous-predictions-based query generator and the dynamic-position-query decoder. Specifically, the previous-predictions-based query generator is designed to separately encode different types of information from previous predictions, which are then effectively utilized by the dynamic-position-query decoder to generate current predictions. Furthermore, we have developed a dual-mode strategy to ensure PrevPredMap's robust performance across both single-frame and temporal modes. Extensive experiments demonstrate that PrevPredMap achieves state-of-the-art performance on the nuScenes and Argoverse2 datasets. Code will be available at https://github.com/pnnnnnnn/PrevPredMap.