Automated Augmentation with Reinforcement Learning and GANs for Robust Identification of Traffic Signs using Front Camera Images

TL;DR

This paper introduces an automated data augmentation framework using reinforcement learning and GANs to improve traffic sign detection across diverse conditions, reducing the need for extensive manual data collection.

Contribution

The work presents a novel end-to-end augmentation method that adapts training data to challenging conditions, enhancing detection accuracy in autonomous driving scenarios.

Findings

Detection precision improved from 0.70 to 0.83 for nighttime images.

Detection recall increased from 0.66 to 0.71 for nighttime images.

Effective mapping of training data across different domains.

Abstract

Traffic sign identification using camera images from vehicles plays a critical role in autonomous driving and path planning. However, the front camera images can be distorted due to blurriness, lighting variations and vandalism which can lead to degradation of detection performances. As a solution, machine learning models must be trained with data from multiple domains, and collecting and labeling more data in each new domain is time consuming and expensive. In this work, we present an end-to-end framework to augment traffic sign training data using optimal reinforcement learning policies and a variety of Generative Adversarial Network (GAN) models, that can then be used to train traffic sign detector modules. Our automated augmenter enables learning from transformed nightime, poor lighting, and varying degrees of occlusions using the LISA Traffic Sign and BDD-Nexar dataset. The proposed method enables mapping training data from one domain to another, thereby improving traffic sign detection precision/recall from 0.70/0.66 to 0.83/0.71 for nighttime images.