High-fidelity performance metrics for generative models in PyTorch

PyTorch - FID calculation with proper image resizing and quantization steps [CVPR 2022]

Pytorch implementation of common image generation metrics.

IS, FID score Pytorch and TF implementation, TF implementation is a wrapper of the official ones.

[CVPR 2024] On the Content Bias in Fréchet Video Distance

Pytorch implementation of Visual DNA, an approach to represent and compare images.

FID computation in Jax/Flax.

Frechet Inception Distance for Keras-based GANs

This Repository Contains Solution to the Assignments of the Generative Adversarial Networks (GANs) Specialization from deeplearning.ai on Coursera Taught by Sharon Zhou, Eda Zhou, Eric Zelikman

Official Repository for the paper "Feature Extraction for Generative Medical Imaging Evaluation: New Evidence Against an Evolving Trend".

A pip-installable evaluator for GANs (IS and FID). Accepts either dataloaders or individual batches. Supports on-the-fly evaluation during training. A working DCGAN SVHN demo script provided.

Lots of evaluation metrics for the generative adversarial networks in pytorch

CXR-ACGAN: Auxiliary Classifier GAN (AC-GAN) for Chest X-Ray (CXR) Images Generation (Pneumonia, COVID-19 and healthy patients) for the purpose of data augmentation. Implemented in TensorFlow, trained on COVIDx CXR-3 dataset.

PyTorch implementation of 'DDPM' (Ho et al., 2020) and training it on CelebA 64×64

Computing the Sliding Fréchet Inception Distance between fake and real images with continous labels

PyTorch implementation of WGAN-GP-based video generation. Includes functionality for measuring Frechet Video Distance and implementing recent research improvements of WGAN-GP. Read paper at https://github.com/talcron/frame-prediction-pytorch/blob/media/paper.pdf

The FID-Evaluator is a tool to analyze how the FID behaves when the embedding space is reduced.

Capturing the special characteristics of Claude Monet's paintings in order to turn ordinary pictures into similar style paintings

Implementation of GAN-based text-to-image models for a comparative study on the CUB and COCO datasets

GAN-based framework to generate depth images of infants from a desired image and pose