Progressive GAN

Progressive Generative Adversarial Networks (ProGANs) are a class of GANs designed to generate high-resolution images by progressively increasing the resolution of both the generator and discriminator networks. Current research focuses on improving ProGAN training efficiency, for example, through techniques like depthwise separable convolutions and incorporating super-resolution GANs, and on enhancing their application in diverse fields such as weather forecasting and medical imaging. ProGANs are proving valuable for generating synthetic data to address data scarcity issues in various domains, particularly where obtaining real data is expensive or ethically challenging, while also enabling novel approaches to inverse problems and adversarial attack detection. The ability to generate high-quality synthetic data has significant implications for training deep learning models and advancing research across numerous scientific disciplines.

Papers

June 14, 2024

Precipitation Nowcasting Using Physics Informed Discriminator Generative Models
Junzhe Yin, Cristian Meo, Ankush Roy, Zeineh Bou Cher, Yanbo Wang, Ruben Imhoff, Remko Uijlenhoet, Justin Dauwels
Physic Informed Neural Network Weather Data Data Driven Nowcasting Progressive GAN

June 7, 2023

A Linearly Convergent GAN Inversion-based Algorithm for Reverse Engineering of Deceptions
Darshan Thaker, Paris Giampouras, René Vidal
Adversarial Attack Practical Algorithm GAN Model GAN Inversion Reverse Engineering Deceptive Diffusion Reliable Deep Learning Progressive GAN

June 5, 2023

Brain tumor segmentation using synthetic MR images -- A comparison of GANs and diffusion models
Muhammad Usman Akbar, Måns Larsson, Anders Eklund
Diffusion Model Generative Adversarial Network Consistent Comparison Brain Tumor Segmentation Synthetic Medical Image Synthetic MRI Progressive GAN

December 27, 2022

General GAN-generated image detection by data augmentation in fingerprint domain
Huaming Wang, Jianwei Fei, Yunshu Dai, Lingyun Leng, Zhihua Xia
Data Augmentation General Text GAN Algorithm Image Detection GAN Image GAN Fingerprint Progressive GAN

March 8, 2022

Regularized Training of Intermediate Layers for Generative Models for Inverse Problems
Sean Gunn, Jorio Cocola, Paul Hand
Generative Model Generative Adversarial Network Inverse Problem Intermediate Layer Inversion Method Progressive GAN

February 24, 2022

Time Efficient Training of Progressive Generative Adversarial Network using Depthwise Separable Convolution and Super Resolution Generative Adversarial Network
Atharva Karwande, Pranesh Kulkarni, Tejas Kolhe, Akshay Joshi, Soham Kamble
Generative Adversarial Network Super Resolution GAN Model Separable Convolution Super Resolution Generative Adversarial Network Progressive GAN

Progressive GAN

Papers

Precipitation Nowcasting Using Physics Informed Discriminator Generative Models

A Linearly Convergent GAN Inversion-based Algorithm for Reverse Engineering of Deceptions

Brain tumor segmentation using synthetic MR images -- A comparison of GANs and diffusion models

General GAN-generated image detection by data augmentation in fingerprint domain

Regularized Training of Intermediate Layers for Generative Models for Inverse Problems

Time Efficient Training of Progressive Generative Adversarial Network using Depthwise Separable Convolution and Super Resolution Generative Adversarial Network