VATLD: A Visual Analytics System to Assess, Understand and Improve Traffic Light Detection
This work addresses the need for thorough assessment of traffic light detectors in safety-critical autonomous driving applications, though it appears incremental as it builds on existing deep learning methods.
The paper tackles the problem of evaluating and improving the accuracy and robustness of traffic light detectors for autonomous driving by proposing VATLD, a visual analytics system that uses disentangled representation learning and semantic adversarial learning, resulting in enhanced performance and actionable insights.
Traffic light detection is crucial for environment perception and decision-making in autonomous driving. State-of-the-art detectors are built upon deep Convolutional Neural Networks (CNNs) and have exhibited promising performance. However, one looming concern with CNN based detectors is how to thoroughly evaluate the performance of accuracy and robustness before they can be deployed to autonomous vehicles. In this work, we propose a visual analytics system, VATLD, equipped with a disentangled representation learning and semantic adversarial learning, to assess, understand, and improve the accuracy and robustness of traffic light detectors in autonomous driving applications. The disentangled representation learning extracts data semantics to augment human cognition with human-friendly visual summarization, and the semantic adversarial learning efficiently exposes interpretable robustness risks and enables minimal human interaction for actionable insights. We also demonstrate the effectiveness of various performance improvement strategies derived from actionable insights with our visual analytics system, VATLD, and illustrate some practical implications for safety-critical applications in autonomous driving.