Road Segmentation Using CNN and Distributed LSTM

TL;DR

This paper introduces a hybrid CNN and distributed LSTM network for road segmentation in autonomous driving, achieving faster processing times while maintaining effective feature extraction.

Contribution

The paper proposes a novel combination of CNN and distributed LSTM layers for efficient road segmentation, reducing computational cost compared to traditional CNN-only models.

Findings

Enhanced feature extraction with combined CNN and LSTM

Reduced processing time compared to pure CNN models

Effective performance on KITTI benchmark

Abstract

In automated driving systems (ADS) and advanced driver-assistance systems (ADAS), an efficient road segmentation is necessary to perceive the drivable region and build an occupancy map for path planning. The existing algorithms implement gigantic convolutional neural networks (CNNs) that are computationally expensive and time consuming. In this paper, we introduced distributed LSTM, a neural network widely used in audio and video processing, to process rows and columns in images and feature maps. We then propose a new network combining the convolutional and distributed LSTM layers to solve the road segmentation problem. In the end, the network is trained and tested in KITTI road benchmark. The result shows that the combined structure enhances the feature extraction and processing but takes less processing time than pure CNN structure.