Robust Traffic Light Detection Using Salience-Sensitive Loss: Computational Framework and Evaluations

TL;DR

This paper introduces a novel traffic light detection model that emphasizes salient lights affecting driver decisions, supported by a new dataset and a salience-sensitive loss function, improving detection recall in autonomous driving.

Contribution

It presents the first US traffic light dataset with salience annotations and a deformable transformer model trained with a salience-sensitive loss for better detection of relevant lights.

Findings

Model trained with Salience-Sensitive Focal Loss shows higher recall.

Introduction of the LAVA Salient Lights Dataset with salience annotations.

Enhanced detection performance for driver-relevant traffic lights.

Abstract

One of the most important tasks for ensuring safe autonomous driving systems is accurately detecting road traffic lights and accurately determining how they impact the driver's actions. In various real-world driving situations, a scene may have numerous traffic lights with varying levels of relevance to the driver, and thus, distinguishing and detecting the lights that are relevant to the driver and influence the driver's actions is a critical safety task. This paper proposes a traffic light detection model which focuses on this task by first defining salient lights as the lights that affect the driver's future decisions. We then use this salience property to construct the LAVA Salient Lights Dataset, the first US traffic light dataset with an annotated salience property. Subsequently, we train a Deformable DETR object detection transformer model using Salience-Sensitive Focal Loss to emphasize stronger performance on salient traffic lights, showing that a model trained with this loss function has stronger recall than one trained without.