Improving diversity of class-conditional generative networks (cGANs) for image classification, using sample reweighting and boosting techniques

Multi-Variate Temporal GAN for Large Scale Video Generation

Class-conditional BigGAN-style training on an ImageNet-10 subset with truncation sweeps and rigorous evaluation (FID, Precision/Recall, PRDC), plus a Gradio demo

Uses a Pretrained Network (using Google's Deepmind) to generate images.

StyleGAN BigGAN for Motorbike Generation

All the computer vision projects and experiments

Data Augmentation optimized for GAN (DAG)

Exploring Generative Adversarial Networks: Improving training techniques

Examples code and images for programmatically generating images using Big Sleep (CLIP + BigGAN).

Creates unique music videos, using a generative adversarial network.

5th place solution for Kaggle Generative Dog Images competition

Using Deep Learning to create fake images of games using PyTorch

Converts signals to images in real time using AI models

Music visualizer using audio and semantic analysis to explore BigGAN latent space.

The Wandering Dreamer: An Synthetic Feedback Loop

Reimplementation of the Paper: Large Scale GAN Training for High Fidelity Natural Image Synthesis

Pytorch implementation of BigGAN Generator with pretrained weights

MichelAI, video clip creator based on BigGAN