Abstract
Lane and road marker segmentation is crucial in autonomous driving, and many related methods have been proposed in this field. However, most of them are based on single-frame prediction, which causes unstable results between frames. Some semantic multi-frame segmentation methods produce error accumulation and are not fast enough. Therefore, we propose a deep learning algorithm that takes into account the continuity information of adjacent image frames, including image sequence processing and an end-to-end trainable multi-input single-output network to jointly process the segmentation of lanes and road markers. In order to emphasize the location of the target with high probability in the adjacent frames and to refine the segmentation result of the current frame, we explicitly consider the time consistency between frames, expand the segmentation region of the previous frame, and use the optical flow of the adjacent frames to reverse the past prediction, then use it as an additional input of the network in training and reasoning, thereby improving the network’s attention to the target area of the past frame. We segmented lanes and road markers on the Baidu Apolloscape lanemark segmentation dataset and CULane dataset, and present benchmarks for different networks. The experimental results show that this method accelerates the segmentation speed of video lanes and road markers by 2.5 times, increases accuracy by 1.4%, and reduces temporal consistency by only 2.2% at most.
Highlights
Unmanned driving technology or auxiliary driving technology has broad development prospects
Most of the lane and road marker segmentation methods are only trained on a single frame of an image, and the correlation between frames is not considered, which leads to unstable segmentation on continuous frame image sequences
A few methods consider the correlation of video sequences, such as methods based on key frames. These methods may cause the cumulative error of semantic segmentation to become increasingly serious as the distance between the current frame and the key frame increases
Summary
Unmanned driving technology or auxiliary driving technology has broad development prospects. The sensing module and the control unit constitute an unmanned driving system [1]. The precondition of stable operation of unmanned driving systems is the understanding and recognition of high performance environments, which depends on the sensing module composed of multiple sensors [2]. The lanes and road markers on the highway do not have special three-dimensional shapes, and collecting the geometric contour of the surrounding environment only by relying on radar is not enough [3]. In addition to radar, visual sensors and computer vision technology are important links in the whole sensing module [4]. Only algorithms that can maintain robustness and achieve high-quality real-time performance under different circumstances and environments are suitable for unmanned driving systems [5]
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.