Enhanced Pix2Pix GAN for Visual Defect Removal in UAV-Captured Images

TL;DR

This paper introduces an enhanced Pix2Pix GAN tailored for removing visual defects from UAV images, demonstrating improved image quality and defect correction through architectural modifications and evaluation on aerial datasets.

Contribution

It presents a novel enhanced Pix2Pix GAN architecture specifically designed for UAV image defect removal, addressing issues like mode collapse and improving image restoration quality.

Findings

Significant improvement in UAV image quality after defect removal

Effective handling of mode collapse in GAN training

Demonstrated success on a custom aerial photograph dataset

Abstract

This paper presents a neural network that effectively removes visual defects from UAV-captured images. It features an enhanced Pix2Pix GAN, specifically engineered to address visual defects in UAV imagery. The method incorporates advanced modifications to the Pix2Pix architecture, targeting prevalent issues such as mode collapse. The suggested method facilitates significant improvements in the quality of defected UAV images, yielding cleaner and more precise visual results. The effectiveness of the proposed approach is demonstrated through evaluation on a custom dataset of aerial photographs, highlighting its capability to refine and restore UAV imagery effectively.