Camera Agnostic Two-Head Network for Ego-Lane Inference
This addresses the limitation of traditional calibration-dependent methods in autonomous driving systems, offering a more flexible solution for varying camera configurations.
The paper tackles the problem of ego-lane inference in autonomous driving by proposing a learning-based method that estimates the ego-lane index from a single image without requiring camera calibration, achieving robust performance validated in diverse environments and camera setups.
Vision-based ego-lane inference using High-Definition (HD) maps is essential in autonomous driving and advanced driver assistance systems. The traditional approach necessitates well-calibrated cameras, which confines variation of camera configuration, as the algorithm relies on intrinsic and extrinsic calibration. In this paper, we propose a learning-based ego-lane inference by directly estimating the ego-lane index from a single image. To enhance robust performance, our model incorporates the two-head structure inferring ego-lane in two perspectives simultaneously. Furthermore, we utilize an attention mechanism guided by vanishing point-and-line to adapt to changes in viewpoint without requiring accurate calibration. The high adaptability of our model was validated in diverse environments, devices, and camera mounting points and orientations.