UniDrive: Towards Universal Driving Perception Across Camera Configurations
This work addresses a practical deployment challenge for autonomous driving systems by improving model adaptability to different sensor setups, though it is incremental as it builds on existing 3D perception methods.
The paper tackles the problem of vision-centric autonomous driving models being sensitive to variations in camera configurations, which hinders deployment across different car models, by proposing UniDrive, a framework that uses unified virtual cameras and ground-aware projection to achieve universal perception, with experiments showing it generalizes to varying configurations with minor performance degradation.
Vision-centric autonomous driving has demonstrated excellent performance with economical sensors. As the fundamental step, 3D perception aims to infer 3D information from 2D images based on 3D-2D projection. This makes driving perception models susceptible to sensor configuration (e.g., camera intrinsics and extrinsics) variations. However, generalizing across camera configurations is important for deploying autonomous driving models on different car models. In this paper, we present UniDrive, a novel framework for vision-centric autonomous driving to achieve universal perception across camera configurations. We deploy a set of unified virtual cameras and propose a ground-aware projection method to effectively transform the original images into these unified virtual views. We further propose a virtual configuration optimization method by minimizing the expected projection error between original and virtual cameras. The proposed virtual camera projection can be applied to existing 3D perception methods as a plug-and-play module to mitigate the challenges posed by camera parameter variability, resulting in more adaptable and reliable driving perception models. To evaluate the effectiveness of our framework, we collect a dataset on CARLA by driving the same routes while only modifying the camera configurations. Experimental results demonstrate that our method trained on one specific camera configuration can generalize to varying configurations with minor performance degradation.