DragTraffic: Interactive and Controllable Traffic Scene Generation for Autonomous Driving
This addresses the need for scalable corner cases in autonomous driving systems, offering an interactive tool for non-experts, though it appears incremental as it adapts existing techniques like DragGAN and diffusion models to a specific domain.
The paper tackles the problem of generating diverse and controllable traffic scenes for autonomous driving evaluation and training by proposing DragTraffic, a framework based on conditional diffusion that allows non-experts to create realistic scenarios, with experiments showing it outperforms existing methods in authenticity, diversity, and freedom.
Evaluating and training autonomous driving systems require diverse and scalable corner cases. However, most existing scene generation methods lack controllability, accuracy, and versatility, resulting in unsatisfactory generation results. Inspired by DragGAN in image generation, we propose DragTraffic, a generalized, interactive, and controllable traffic scene generation framework based on conditional diffusion. DragTraffic enables non-experts to generate a variety of realistic driving scenarios for different types of traffic agents through an adaptive mixture expert architecture. We employ a regression model to provide a general initial solution and a refinement process based on the conditional diffusion model to ensure diversity. User-customized context is introduced through cross-attention to ensure high controllability. Experiments on a real-world driving dataset show that DragTraffic outperforms existing methods in terms of authenticity, diversity, and freedom. Demo videos and code are available at https://chantsss.github.io/Dragtraffic/.