Lane Detection in Low-light Conditions Using an Efficient Data Enhancement : Light Conditions Style Transfer
This work addresses the problem of reliable lane detection for autonomous vehicles in challenging low-light environments, representing an incremental improvement over existing methods.
The paper tackles lane detection in low-light conditions by proposing a style-transfer-based data enhancement method using GANs to generate low-light images, which improves the lane detector's adaptability without requiring extra annotations or inference overhead, as validated on the CULane benchmark with ERFNet.
Nowadays, deep learning techniques are widely used for lane detection, but application in low-light conditions remains a challenge until this day. Although multi-task learning and contextual-information-based methods have been proposed to solve the problem, they either require additional manual annotations or introduce extra inference overhead respectively. In this paper, we propose a style-transfer-based data enhancement method, which uses Generative Adversarial Networks (GANs) to generate images in low-light conditions, that increases the environmental adaptability of the lane detector. Our solution consists of three parts: the proposed SIM-CycleGAN, light conditions style transfer and lane detection network. It does not require additional manual annotations nor extra inference overhead. We validated our methods on the lane detection benchmark CULane using ERFNet. Empirically, lane detection model trained using our method demonstrated adaptability in low-light conditions and robustness in complex scenarios. Our code for this paper will be publicly available.