VPGNet: Vanishing Point Guided Network for Lane and Road Marking Detection and Recognition

Abstract

In this paper, we propose a unified end-to-end trainable multi-task network that jointly handles lane and road marking detection and recognition that is guided by a vanishing point under adverse weather conditions. We tackle rainy and low illumination conditions, which have not been extensively studied until now due to clear challenges. For example, images taken under rainy days are subject to low illumination, while wet roads cause light reflection and distort the appearance of lane and road markings. At night, color distortion occurs under limited illumination. As a result, no benchmark dataset exists and only a few developed algorithms work under poor weather conditions. To address this shortcoming, we build up a lane and road marking benchmark which consists of about 20,000 images with 17 lane and road marking classes under four different scenarios: no rain, rain, heavy rain, and night. We train and evaluate several versions of the proposed multi-task network and validate the importance of each task. The resulting approach, VPGNet, can detect and classify lanes and road markings, and predict a vanishing point with a single forward pass. Experimental results show that our approach achieves high accuracy and robustness under various conditions in realtime (20 fps). The benchmark and the VPGNet model will be publicly available 1 .

HoughNet: Neural Network Architecture for Vanishing Points Detection

“…Distance d 1 represents the average deviation from 90 • of the document rectangle regardless its orientation (10). Distance d 2 allows us to estimate how good we managed to correct the orientation of the document (11).…”

“…The neural network yields two images, one for the vertical vanishing point and one for the horizontal vanishing point. From every image we take a point with maximum intensity as an answer and transform its coordinates back to the original image coordinates space using equations (8) and (9) mixed with coordinates transform according to convolution layers.…”

Generating All the Roads to Rome: Road Layout Randomization for Improved Road Marking Segmentation

“…The first large-scale semantic road marking dataset was recently introduced in [17], however it is extremely expensive to manually expand this to all environments and conditions. Road marking segmentation as demonstrated in [4] is closest to the application of this paper. The authors train a network for semantic road marking segmentation and improve their results by predicting the vanishing point simultaneously.…”

“…Retrieving more examples is labour intensive in terms of driving and labelling time. Consequently, trained classifiers show decreased performance on infrequently-occurring classes [4].…”

Multi-lane Detection Using Instance Segmentation and Attentive Voting

“…This also reduces the need to adequately capture the background diversity and dependencies, such as traffic and geometric constraints. For example, VPGNet [16] predicts vanishing point on the input frame for guiding lane and road marking detection and recognition in adverse circumstances this introduces more training data to capture the input diversity for vanishing point prediction and also vanishing point works well on straight roads but not on curving roads or roundabouts.…”