DriveSceneGen: Generating Diverse and Realistic Driving Scenarios from Scratch
This addresses the data scarcity issue for autonomous driving system development, though it is an incremental advancement in scenario generation.
The paper tackles the problem of insufficient quantity and diversity in real-world driving datasets for autonomous driving by introducing DriveSceneGen, a method that generates diverse and realistic driving scenarios from scratch, with experimental results on 5k scenarios showing high fidelity and diversity.
Realistic and diverse traffic scenarios in large quantities are crucial for the development and validation of autonomous driving systems. However, owing to numerous difficulties in the data collection process and the reliance on intensive annotations, real-world datasets lack sufficient quantity and diversity to support the increasing demand for data. This work introduces DriveSceneGen, a data-driven driving scenario generation method that learns from the real-world driving dataset and generates entire dynamic driving scenarios from scratch. DriveSceneGen is able to generate novel driving scenarios that align with real-world data distributions with high fidelity and diversity. Experimental results on 5k generated scenarios highlight the generation quality, diversity, and scalability compared to real-world datasets. To the best of our knowledge, DriveSceneGen is the first method that generates novel driving scenarios involving both static map elements and dynamic traffic participants from scratch.