Generative Machine Learning GAN
The dataset was web-scraped for an original 20k samples, then a custom MRCNN model was trained for image segmentation and cropping before being fed into the 128 DCGAN, trained on local hardware, 1660
Generative Model Course
[Deep Learning Specialization DeepLearning.AI](https://www.deeplearning.ai/program/deep-learning-specialization/)
[Machine Learning for Musicians Berklee](https://college.berklee.edu/courses/mtec-345)
- MUSIC DATA MINING - a Music Information Retrieval (MIR) Online Course at Kadenze
[Arts and Entertainment Technology - Online Course Kadenze](https://www.kadenze.com/courses/foundations-of-arts-and-entertainment-technologies-i/info)
- Introduction to Generative Arts and Computational Creativity - an Online Course at Kadenze
- UAL - Apply Creative Machine Learning - Institute of CodingInstitute of Coding
- Machine Learning for Musicians and Artists - an Online Machine Art Course at Kadenze
- Artificial Images - YouTube
- NVlabs/stylegan3: Official PyTorch implementation of StyleGAN3
[AI Demos NVIDIA Research](https://www.nvidia.com/en-us/research/ai-demos/) GAN
- HyperStyle StyleGAN Inversion
- Making Anime Faces With StyleGAN · Gwern.net
- cedricoeldorf/ConditionalStyleGAN: Conditional implementation for NVIDIA's StyleGAN architecture
- NVlabs/stylegan: StyleGAN - Official TensorFlow Implementation
- StyleGAN versions
- t04glovern/stylegan-pokemon: Generating Pokemon cards using a mixture of StyleGAN and RNN to create beautiful & vibrant cards ready for battle!
- 2011.05552End-to-End Chinese Landscape Painting Creation Using Generative Adversarial Networks
- 1812.04948 A Style-Based Generator Architecture for Generative Adversarial Networks
GAN Image Superresolution
- junyanz/CycleGAN: Software that can generate photos from paintings, turn horses into zebras, perform style transfer, and more.
- kaonashi-tyc/zi2zi: Learning Chinese Character style with conditional GAN
- rosinality/style-based-gan-pytorch: Implementation A Style-Based Generator Architecture for Generative Adversarial Networks in PyTorch
- taki0112/StyleGAN-Tensorflow: Simple & Intuitive Tensorflow implementation of StyleGAN (CVPR 2019 Oral)
- mtobeiyf/sketch-to-art: 🖼 Create artwork from your casual sketch with GAN and style transfer
- taki0112/UGATIT: Official Tensorflow implementation of U-GAT-IT: Unsupervised Generative Attentional Networks with Adaptive Layer-Instance Normalization for Image-to-Image Translation (ICLR 2020)
- heavenstobetsy/ArtGenerationwithStyleGan: Fauvist art generation using StyleGAN
- zhenxuan00/triple-gan: See Triple-GAN-V2 in PyTorch: https://github.com/taufikxu/Triple-GAN
- Mawiszus/TOAD-GAN: Official repository for "TOAD-GAN: Coherent Style Level Generation from a Single Example" by Maren Awiszus, Frederik Schubert and Bodo Rosenhahn.
- schrum2/GameGAN: Interactive GAN evolution of Mario and Zelda levels.
- changebo/HCCG-CycleGAN: Handwritten Chinese Characters Generation
- An enhanced 3D model and generative adversarial network for automated generation of horizontal building mask images and cloudless aerial photographs - ScienceDirect
- Realless Generative webs with blinking eyes
- deepjazz: deep learning for jazz jisungk/deepjazz: Deep learning driven jazz generation using Keras & Theano!
- KarthikNayak/DeepRock: Rock Music using Deep Learning
- salu133445/musegan: An AI for Music Generation reddit
- Music Transformer: Generating Music with Long-Term Structure
- Generative Deep Learning for Virtuosic Classical Music: Generative Adversarial Networks as Renowned Composers : MachineLearning
- Real-time Performance RNN in the Browser
- GRUV: Algorithmic Music Generation using Recurrent Neural Networks - Aran Nayebi
- Hyperchroma A music player that creates real-time music videos (hyperchroma.app)
- Pix2pix vs CycleGAN
- Cycle-GAN can work in an ‘unpaired' manner and various architectural differences. Unpaired image translation is much harder, as you demand the model to learn objects of different shapes, size, angle, texture, location in different scenes and settings on top of the actual task (coloring in this case). Requires more data and you don't have fine control on the learning. Formulating the coloring problem as a paired task makes more sense as you simply decrease the complexity of the problem without increasing data collection/annotation work.
- The whole point about using CycleGAN is that it can learn in unpaired situations. And it works well in the context of style transfer tasks where the changes are really bold and less nuanced. But, in the context of image colorization, the changes are really subtle and also there are way more options to choose colors than changing a horse to zebra. The other thing is that learning to change a colored image to black and white is much easier for the model than learning to colorize it which can lead to a bad learning procedure.
- The most prominent differences is that CycleGAN helps when you have unpaired images and you want to go from one class to the other (Horse to Zebra for example) but in the Pix2Pix paper, the images that you get after the inference, are the input images but with some new features (black&white to colorized or day time to night time of a scene). In pix2pix, a conditional GAN (one generator and one discriminator) is used with some supervision from L1 loss. In CycleGAN, you need two generators and two discriminators to do the task: one pair for going from class A to B and one pair for going from class B to A. Also you need Cycle Consistency Loss to make sure that the models learn to undo the changes they make.
- pix2pixhd is pix2pix in higher resolution
- pix2pix tutorial and example:
Machine Learning: Image Generator
- saharmor/dalle-playground: A playground to generate images from any text prompt using DALL-E Mini and based on OpenAI's DALL-E https://openai.com/blog/dall-e/