A Brief Survey and an Application of Semantic Image Segmentation for Autonomous Driving

TL;DR

This paper reviews deep learning methods, specifically Fully Convolutional Networks, for semantic image segmentation to enhance autonomous vehicle perception, comparing four architectures through experiments and visualizations.

Contribution

It applies and compares four FCN architectures for semantic segmentation in autonomous driving, demonstrating their effectiveness and visualization of segmentation accuracy.

Findings

FCN architectures achieve high segmentation accuracy

Visualizations show differences in segmentation precision among models

Experimental results favor certain FCN variants for autonomous driving

Abstract

Deep learning is a fast-growing machine learning approach to perceive and understand large amounts of data. In this paper, general information about the deep learning approach which is attracted much attention in the field of machine learning is given in recent years and an application about semantic image segmentation is carried out in order to help autonomous driving of autonomous vehicles. This application is implemented with Fully Convolutional Network (FCN) architectures obtained by modifying the Convolutional Neural Network (CNN) architectures based on deep learning. Experimental studies for the application are utilized 4 different FCN architectures named FCN-AlexNet,FCN-8s, FCN-16s and FCN-32s. For the experimental studies, FCNs are first trained separately and validation accuracies of these trained network models on the used dataset is compared. In addition, image segmentation inferences are visualized to take account of how precisely FCN architectures can segment objects.