projected-gan
joliGEN
| projected-gan | joliGEN | |
|---|---|---|
| 8 | 4 | |
| 878 | 201 | |
| 0.0% | 3.5% | |
| 0.0 | 9.4 | |
| about 2 years ago | 3 days ago | |
| Python | Python | |
| MIT License | GNU General Public License v3.0 or later |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
projected-gan
-
How to generate Image Morphing Animation with Custom images as keyframes?
Use the images as a training dataset for this
-
[D] Is the GAN architecture currently old-fashioned?
If you are looking for more traditional noise -> xxx GANs, go for https://github.com/autonomousvision/projected_gan/. Another recent work is https://github.com/nupurkmr9/vision-aided-gan.
- Any idea what’s going on here? Why are the losses spiking all the sudden?
-
Generated deep space objects
Made with two ml algorithms, one that generates a tiny 256x256 'base' image, and another that upscales it. Both of them have been trained on a astrophotography-based dataset, but the model architecture are open sourced from Projected-Gan and Real-ESRGAN respectively.
- Pokemon GAN
-
[R][P] Projected GANs Converge Faster, Pokemon + Hugging Face Spaces Demo
github: https://github.com/autonomousvision/projected_gan
-
Have image generation project idea, need tool pointers though
Projected GANs are maybe the currently best model for small datasets. You could also check out StyleGAN 2 ada. https://github.com/autonomousvision/projected_gan
-
Researchers Propose ‘Projected-GANs’, To Improve Image Quality, Sample Efficiency, And Convergence Speed
Researchers from the University of Tübingen, Max Planck Institute for Intelligent Systems, and Heidelberg have studied ways to improve GAN training by using pre-trained representations. The researchers proposed a more effective strategy (Projected-GAN) that combines features across channels and resolutions.
joliGEN
-
[D] Question about using diffusion to denoise images
Absolutely, I do second this, Palette is what you are looking for. We have a modified version in JoliGAN, with PR for various conditioning, including masks and sketches, cf https://github.com/jolibrain/joliGAN/pull/339
-
[D] Is the GAN architecture currently old-fashioned?
We use https://github.com/jolibrain/joliGAN which is a lib for image2image with additional "semantic" constraints. I.e. when there's a need to conserve labels, physics, anything between the two domains. This lib aggregates and improves on existing works.
-
[P] Real-time AR for jewelry virtual try on that looks real, done with joliGAN, based on a few 2D videos and no 3D model
We thought we'd share some technical details since the underlying code, JoliGAN is Open Source, https://github.com/jolibrain/joliGAN
-
[D] Augmentation in GAN
Look at DiffAug, deceive, and such. It's all implemented into Joligan, https://github.com/jolibrain/joliGAN We're able to train with small datasets, though not always optimally. With large datasets results are outstanding.
What are some alternatives?
Real-ESRGAN - Real-ESRGAN aims at developing Practical Algorithms for General Image/Video Restoration.
hifigan-denoiser - HiFi-GAN: High Fidelity Denoising and Dereverberation Based on Speech Deep Features in Adversarial Networks
taming-transformers - Taming Transformers for High-Resolution Image Synthesis
PassGAN - A Deep Learning Approach for Password Guessing (https://arxiv.org/abs/1709.00440)
vision-aided-gan - Ensembling Off-the-shelf Models for GAN Training (CVPR 2022 Oral)
cycle-gan-pytorch - This repository contains an implementation of the Cylce-GAN architecture for style transfer along with instructions to train on an own dataset.
dnn_from_scratch - A high level deep learning library for Convolutional Neural Networks,GANs and more, made from scratch(numpy/cupy implementation).
deepdetect - Deep Learning API and Server in C++14 support for Caffe, PyTorch,TensorRT, Dlib, NCNN, Tensorflow, XGBoost and TSNE
style-aware-discriminator - CVPR 2022 - Official PyTorch implementation of "A Style-Aware Discriminator for Controllable Image Translation"
nni - An open source AutoML toolkit for automate machine learning lifecycle, including feature engineering, neural architecture search, model compression and hyper-parameter tuning.
ALAE - [CVPR2020] Adversarial Latent Autoencoders