A System-driven Automatic Ground Truth Generation Method for DL Inner-City Driving Corridor Detectors

TL;DR

This paper introduces an automated ground truth generation method for semantic segmentation in autonomous driving, significantly reducing manual labeling effort and enabling continuous perception system improvement.

Contribution

It presents a system-driven approach that automates ground truth labeling for drivable corridor detection, drastically decreasing manual effort and facilitating automated data loops.

Findings

Reduces manual labeling effort by over 150 times.

Enables continuous perception module improvement.

Applicable to high-bandwidth sensor data.

Abstract

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.