English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

After the 2017 TuSimple Lane Detection Challenge, its evaluation based onaccuracy and F1 score has become the de facto standard to measure theperformance of lane detection methods. In this work, we conduct the firstlarge-scale empirical study to evaluate the robustness of state-of-the-art lanedetection methods under physical-world adversarial attacks in autonomousdriving. We evaluate 4 major types of lane detection approaches with theconventional evaluation and end-to-end evaluation in autonomous drivingscenarios and then discuss the security proprieties of each lane detectionmodel. We demonstrate that the conventional evaluation fails to reflect therobustness in end-to-end autonomous driving scenarios. Our results show thatthe most robust model on the conventional metrics is the least robust in theend-to-end evaluation. Although the competition dataset and its metrics haveplayed a substantial role in developing performant lane detection methods alongwith the rapid development of deep neural networks, the conventional evaluationis becoming obsolete and the gap between the metrics and practicality iscritical. We hope that our study will help the community make further progressin building a more comprehensive framework to evaluate lane detection models.

On Robustness of Lane Detection Models to Physical-World Adversarial Attacks in Autonomous Driving