Oct 12, 2017 (modified: Oct 12, 2017)NIPS 2017 Workshop MLITS Submissionreaders: everyone
Abstract:In this paper, we investigate the robustness of traffic sign recognition algorithms under challenging conditions. Existing datasets are limited in terms of their size and challenging condition coverage, which motivated us to generate the Challenging Unreal and Real Environments for Traffic Sign Recognition (CURE-TSR) dataset. It includes more than two million traffic sign images that are based on real-world and simulator data. We benchmark the performance of existing solutions in real-world scenarios and analyze the performance variation with respect to challenging conditions. We show that challenging conditions can decrease the performance of baseline methods significantly, especially if these challenging conditions result in loss or misplacement of spatial information. We also investigate the utilization of simulator data along with real-world data and show that hybrid training can enhance the average recognition performance in real-world scenarios.
TL;DR:We investigate the robustness of traffic sign recognition algorithms under challenging conditions and utilize simulator data along with real-world data to enhance recognition performance in real-world scenarios.
Keywords:Traffic sign recognition, traffic sign dataset, performance evaluation under challenging conditions, machine learning for autonomous vehicles, simulator and real-world data
Enter your feedback below and we'll get back to you as soon as possible.