Application of Ghost-DeblurGAN to Fiducial Marker Detection

TL;DR

This paper introduces Ghost-DeblurGAN, a lightweight real-time deblurring network, and a new dataset YorkTag, to improve fiducial marker detection in motion-blurred images for robotic applications.

Contribution

The paper presents Ghost-DeblurGAN, a novel lightweight GAN for real-time motion deblurring, and introduces YorkTag, a large-scale dataset for benchmarking deblurring in fiducial marker detection.

Findings

Ghost-DeblurGAN significantly improves marker detection accuracy.

YorkTag dataset enables effective training and evaluation of deblurring models.

The approach is suitable for real-time robotic applications.

Abstract

Feature extraction or localization based on the fiducial marker could fail due to motion blur in real-world robotic applications. To solve this problem, a lightweight generative adversarial network, named Ghost-DeblurGAN, for real-time motion deblurring is developed in this paper. Furthermore, on account that there is no existing deblurring benchmark for such task, a new large-scale dataset, YorkTag, is proposed that provides pairs of sharp/blurred images containing fiducial markers. With the proposed model trained and tested on YorkTag, it is demonstrated that when applied along with fiducial marker systems to motion-blurred images, Ghost-DeblurGAN improves the marker detection significantly. The datasets and codes used in this paper are available at: https://github.com/York-SDCNLab/Ghost-DeblurGAN.