A System-driven Automatic Ground Truth Generation Method for DL Inner-City Driving Corridor Detectors
This addresses a critical bottleneck in automated driving systems by enabling more efficient and scalable data labeling for perception tasks, though it is incremental as it builds on existing supervised learning methods.
The paper tackles the high manual effort and cost of generating ground truth data for perception modules in automated driving by proposing an automatic labeling approach for semantic segmentation of drivable corridors, reducing manual effort by a factor of 150 or more.
Data-driven perception approaches are well-established in automated driving systems. In many fields even super-human performance is reached. Unlike prediction and planning approaches, mainly supervised learning algorithms are used for the perception domain. Therefore, a major remaining challenge is the efficient generation of ground truth data. As perception modules are positioned close to the sensor, they typically run on raw sensor data of high bandwidth. Due to that, the generation of ground truth labels typically causes a significant manual effort, which leads to high costs for the labelling itself and the necessary quality control. In this contribution, we propose an automatic labeling approach for semantic segmentation of the drivable ego corridor that reduces the manual effort by a factor of 150 and more. The proposed holistic approach could be used in an automated data loop, allowing a continuous improvement of the depending perception modules.