SHIFT: A Synthetic Driving Dataset for Continuous Multi-Task Domain Adaptation
This dataset addresses the challenge of domain adaptation for autonomous driving perception systems, though it is incremental as it builds on existing synthetic data efforts.
The authors tackled the problem of autonomous driving systems adapting to continuously changing environments by introducing SHIFT, the largest multi-task synthetic dataset, which includes discrete and continuous shifts in weather, time, and traffic conditions, enabling the study of performance degradation and development of adaptation strategies.
Adapting to a continuously evolving environment is a safety-critical challenge inevitably faced by all autonomous driving systems. Existing image and video driving datasets, however, fall short of capturing the mutable nature of the real world. In this paper, we introduce the largest multi-task synthetic dataset for autonomous driving, SHIFT. It presents discrete and continuous shifts in cloudiness, rain and fog intensity, time of day, and vehicle and pedestrian density. Featuring a comprehensive sensor suite and annotations for several mainstream perception tasks, SHIFT allows investigating the degradation of a perception system performance at increasing levels of domain shift, fostering the development of continuous adaptation strategies to mitigate this problem and assess model robustness and generality. Our dataset and benchmark toolkit are publicly available at www.vis.xyz/shift.