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

TL;DR

A neural network is presented that effectively removes visual defects from UAV-captured images, specifically engineered to address visual defects in UAV imagery, using an enhanced Pix2Pix GAN.

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.

Figures (6)

• Figure 1: Sample images (512×512 pixels)
• Figure 2: Degraded images
• Figure 3: (Left) Input image, (Center) Enhanced by baseline Pix2Pix GAN, (Right) Enhanced by Pix2Pix GAN with the proposed method.
• Figure 4: Discriminator loss dynamics over epochs.
• Figure 5: FID score dynamics over epochs.