Description:	Adverse weather conditions pose a significant challenge to the widespread adoption of Autonomous Vehicles (AVs) by impacting sensors like LiDARs and cameras. Even though Collaborative Perception (CP) improves AV perception in difficult conditions, existing CP datasets lack adverse weather conditions. To address this, we introduce Adver-City, the first open-source synthetic CP dataset focused on adverse weather conditions. Simulated in CARLA with OpenCDA, it contains over 24 thousand frames, over 890 thousand annotations, and 110 unique scenarios across six different weather conditions: clear weather, soft rain, heavy rain, fog, foggy heavy rain and, for the first time in a synthetic CP dataset, glare. It has six object categories including pedestrians and cyclists, and uses data from vehicles and roadside units featuring LiDARs, RGB and semantic segmentation cameras, GNSS, and IMUs. Its scenarios, based on real crash reports, depict the most relevant road configurations for adverse weather and poor visibility conditions, varying in object density, with both dense and sparse scenes, allowing for novel testing conditions of CP models. Benchmarks run on the dataset show that weather conditions created challenging conditions for perception models, reducing multi-modal object detection performance by up to 19%, while object density affected LiDAR-based detection by up to 29%. The code and documentation are available at https://labs.cs.queensu.ca/quarrg/datasets/adver-city/.
Authors:	Karvat, Mateus; Queen's University; 0000-0001-5196-7228
Keywords:	autonomous vehicles self-driving cars collaborative perception lidar camera gnss imu adverse weather conditions multi-agent carla simulated data
Field of Research:	Computer and information sciences > Artificial intelligence (AI) > Computer vision in artificial intelligence
Publication Date:	2024-10-31
Publisher:	Federated Research Data Repository / dépôt fédéré de données de recherche
Funder:	Natural Sciences and Engineering Research Council; NSERC Discovery
URI:	https://doi.org/10.20383/103.01084
Related Identifiers:	This dataset is part/subset of https://arxiv.org/abs/2410.06380 This dataset is compiled/created by https://proceedings.mlr.press/v78/dosovitskiy17a/dosovitskiy17a.pdf This dataset is compiled/created by https://doi.org/10.48550/arXiv.2107.06260 This dataset is derived from https://github.com/QUARRG/Adver-City