Implementation of augmentation, SMOTE, ADASYN, AE, DGAN for oversampling image data used in detection (YOLO and others one stage detectors). Maritime flags/code scenario.

A comprehensive implementation of GAN architectures including DCGAN, WGAN-GP, Pix2Pix, and CycleGAN for image generation and domain translation. This project explores tackling mode collapse and performing paired and unpaired image-to-image synthesis using PyTorch

Celeb Faces Generator using DGANS and CNN in Pytorch