CURE-TSR: Challenging Unreal and Real Environments for Traffic Sign Recognition

By Dogancan Temel, Gukyeong Kwon, M. Prabhushankar, Ghassan Al-Regib. 2017

81 citations recorded by the source corpus.

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 effect of data augmentation and show that utilization of simulator data along with real-world data enhance the average recognition performance in real-world scenarios. The dataset is publicly available at this https URL

Subfields: Civil and Structural Engineering, Computer Vision and Pattern Recognition, Evaluation & Benchmarks, Human-Computer Interaction

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Civil and Structural Engineering

1. Dogancan Temel
2. Ghassan Al-Regib
3. Thomas G. Dietterich
4. Dan Hendrycks

In Computer Vision and Pattern Recognition

1. Dan Hendrycks
2. Thomas G. Dietterich
3. Ghassan Al-Regib
4. Dogancan Temel
5. Jinsol Lee
6. Luc Van Gool
7. Trevor Darrell
8. Chen Change Loy

In Evaluation & Benchmarks

1. Dan Hendrycks
2. Thomas G. Dietterich
3. Dogancan Temel
4. Ghassan Al-Regib
5. Jinsol Lee
6. Balaji Lakshminarayanan
7. Dustin Tran
8. Neel Guha

In Human-Computer Interaction

1. Dogancan Temel
2. Ghassan Al-Regib

Citations in the corpus

• Benchmarking Neural Network Robustness to Common Corruptions and Perturbations (2019, background): “Temel et al. (2017; 2018); Temel & AlRegib (2018) propose different corrupted datasets for object and traffic sign recognition.”
• CURE-OR: Challenging Unreal and Real Environments for Object Recognition (2018, background): “Recent studies started investigating the effect of challenging conditions in images in terms of recognition performance [15– 20].”
• Traffic Signs in the Wild: Highlights from the IEEE Video and Image Processing Cup 2017 Student Competition [SP Competitions] (2018, methodology): “To address such shortcomings, the Challenging Unreal and Real Environments for Traffic Sign Detection (CURE-TSD) dataset was recently introduced [1], which was also utilized for traffic sign recognition in [2].”