Deep Semantic Segmentation for Automated Driving: Taxonomy, Roadmap and Challenges
This is an incremental survey paper addressing the adaptation of semantic segmentation for automated driving systems, targeting researchers and practitioners in autonomous vehicles.
The paper explores semantic segmentation for automated driving, highlighting that current algorithms are generic and lack integration with driving-specific goals, and it empirically evaluates architectures on the CamVid dataset for accuracy and speed.
Semantic segmentation was seen as a challenging computer vision problem few years ago. Due to recent advancements in deep learning, relatively accurate solutions are now possible for its use in automated driving. In this paper, the semantic segmentation problem is explored from the perspective of automated driving. Most of the current semantic segmentation algorithms are designed for generic images and do not incorporate prior structure and end goal for automated driving. First, the paper begins with a generic taxonomic survey of semantic segmentation algorithms and then discusses how it fits in the context of automated driving. Second, the particular challenges of deploying it into a safety system which needs high level of accuracy and robustness are listed. Third, different alternatives instead of using an independent semantic segmentation module are explored. Finally, an empirical evaluation of various semantic segmentation architectures was performed on CamVid dataset in terms of accuracy and speed. This paper is a preliminary shorter version of a more detailed survey which is work in progress.