ENet-21: An Optimized light CNN Structure for Lane Detection
Seyed Rasoul Hosseini, Hamid Taheri, Mohammad Teshnehlab
TL;DR
The paper tackles lane detection for autonomous driving where the number of lanes varies and lane changes occur. It presents an end-to-end approach using a lightweight ENet-based CNN to jointly perform semantic segmentation and predict horizontal and vertical affinity fields, which cluster lane pixels into distinct lanes during decoding. Key contributions include a compact backbone with three heads for segmentation and affinity outputs, a row-wise clustering mechanism via HAF and VAF, and competitive TuSimple results with only 0.25 million parameters and 3.14 GFLOPs. The approach yields high F1 scores and low false positives, demonstrating practical potential for real-time lane detection on resource-constrained platforms.
Abstract
Lane detection for autonomous vehicles is an important concept, yet it is a challenging issue of driver assistance systems in modern vehicles. The emergence of deep learning leads to significant progress in self-driving cars. Conventional deep learning-based methods handle lane detection problems as a binary segmentation task and determine whether a pixel belongs to a line. These methods rely on the assumption of a fixed number of lanes, which does not always work. This study aims to develop an optimal structure for the lane detection problem, offering a promising solution for driver assistance features in modern vehicles by utilizing a machine learning method consisting of binary segmentation and Affinity Fields that can manage varying numbers of lanes and lane change scenarios. In this approach, the Convolutional Neural Network (CNN), is selected as a feature extractor, and the final output is obtained through clustering of the semantic segmentation and Affinity Field outputs. Our method uses less complex CNN architecture than existing ones. Experiments on the TuSimple dataset support the effectiveness of the proposed method.