GenerativeModels

GenerativeModels copied to clipboard

Add tutorial with approach similar to eDiff-I (https://arxiv.org/pdf/2211.01324.pdf) which relies on an ensemble of expert denoisers.

Warvito

Add in [CONTRIBUTING.md](https://github.com/Project-MONAI/GenerativeModels/blob/main/CONTRIBUTING.md) instructions about how to create and run a tuorial using jupytext.

Warvito

A possible new feature to support described in https://github.com/AUTOMATIC1111/stable-diffusion-webui/pull/5521 and https://github.com/huggingface/diffusers/pull/2059 . Instead upscaling atthe image level, we can perform upscaling at the latent space level. This might help to...

Warvito

With the transformer and the VQ-VAE, I believe we have most of the elements to implement MUSE (https://muse-model.github.io/) from "Muse: Text-To-Image Generation via Masked Generative Transformers".

Warvito

Add cycle consistent embedding GAN from https://openreview.net/forum?id=jgBzGIG-kB

Warvito

@danieltudosiu and I have been discussing possibly having a class for the feature extractors that the Perceptual loss and the FID metric could use. It would include the mechanisms to...

Warvito

Available Datasets

6

Let's list the available datasets that we have access to. * VerSe Dataset https://github.com/anjany/verse (labels are questionable) * ADNI * Decathlon

ericspod

Tutorial about adaptative weights for adversarial training

1

In the original code from taming transformers and latent diffusion models, the weight for the discriminator adversarial loss is defined by the ratio between the nll_loss and the g_loss gradients...

Warvito

Currently cannot tests distributed tests in my test environment. If somebody have access to multiple GPU, please, try to create an unit test to evaluate distributed_synchronization method from EMAQuantizer

Warvito

Add components necessary to perform super-resolution based on GANs.

Warvito