Road Segmentation for ADAS/AD Applications

TL;DR

This paper compares the impact of model architecture and dataset choice on road segmentation accuracy for autonomous driving, demonstrating that a modified VGG-16 outperforms U-Net across datasets despite fewer training epochs.

Contribution

It provides an empirical comparison of VGG-16 and U-Net architectures on different datasets, highlighting architecture and dataset effects on segmentation performance.

Findings

VGG-16 outperforms U-Net in cross-dataset testing.

Both models achieve high accuracy on their respective datasets.

Model architecture and dataset choice significantly influence segmentation results.

Abstract

Accurate road segmentation is essential for autonomous driving and ADAS, enabling effective navigation in complex environments. This study examines how model architecture and dataset choice affect segmentation by training a modified VGG-16 on the Comma10k dataset and a modified U-Net on the KITTI Road dataset. Both models achieved high accuracy, with cross-dataset testing showing VGG-16 outperforming U-Net despite U-Net being trained for more epochs. We analyze model performance using metrics such as F1-score, mean intersection over union, and precision, discussing how architecture and dataset impact results.