AID4AD: Aerial Image Data for Automated Driving Perception

2025-08-04

AI Summary

To address the challenges of unavailable, outdated, or prohibitively expensive high-definition (HD) maps, this paper proposes leveraging georegistered aerial imagery as scalable environmental context for autonomous driving perception. We introduce AID4AD—the first aerial image dataset precisely aligned with the nuScenes coordinate system—constructed via SLAM point-cloud–assisted registration, lens distortion correction, and rigorous human-verified ground truth annotation, significantly improving spatial alignment accuracy. AID4AD enables research on automated aerial image registration and validates the feasibility of HD-map–free environmental modeling. Experiments demonstrate that incorporating AID4AD improves online map construction accuracy by 15–23% and trajectory prediction performance by 2%. The dataset, models, and source code are publicly released.

Technology Category

• Computer Vision: Remote Sensing / Geospatial AI
• Intelligent Robots: Localization, Mapping, and Navigation
• Application Domains: Mobility, Driving & Flight

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

This work investigates the integration of spatially aligned aerial imagery into perception tasks for automated vehicles (AVs). As a central contribution, we present AID4AD, a publicly available dataset that augments the nuScenes dataset with high-resolution aerial imagery precisely aligned to its local coordinate system. The alignment is performed using SLAM-based point cloud maps provided by nuScenes, establishing a direct link between aerial data and nuScenes local coordinate system. To ensure spatial fidelity, we propose an alignment workflow that corrects for localization and projection distortions. A manual quality control process further refines the dataset by identifying a set of high-quality alignments, which we publish as ground truth to support future research on automated registration. We demonstrate the practical value of AID4AD in two representative tasks: in online map construction, aerial imagery serves as a complementary input that improves the mapping process; in motion prediction, it functions as a structured environmental representation that replaces high-definition maps. Experiments show that aerial imagery leads to a 15-23% improvement in map construction accuracy and a 2% gain in trajectory prediction performance. These results highlight the potential of aerial imagery as a scalable and adaptable source of environmental context in automated vehicle systems, particularly in scenarios where high-definition maps are unavailable, outdated, or costly to maintain. AID4AD, along with evaluation code and pretrained models, is publicly released to foster further research in this direction: https://github.com/DriverlessMobility/AID4AD.

Problem

• Integrate aerial imagery into automated vehicle perception tasks
• Align high-resolution aerial images with local vehicle coordinates
• Improve map construction and motion prediction using aerial data

Innovation

• SLAM-based aerial and ground data alignment
• High-resolution aerial imagery integration
• Alignment workflow for spatial fidelity