Generative Adversarial Networks Exercises

The exercises include Stable Diffusion, Style Transfer, Style GAN, Image Inpainting, Virtual Try On, Talking Head, Music Seperation, VoiceSplit.

---

Stable Diffusion

Kaggle: https://www.kaggle.com/code/rkuo2000/stable-diffusion/

Demo:

• Img2Img Stable Diffusion

• Best Stable Diffusion Prompts

• Turn face into pixar character:

---

Style Transfer

Kaggle: rkuo2000/neural-style

---

StyleGAN2

Kaggle: rkuo2000/stylegan2-ada

pix2style2pix

Kaggle: rkuo2000/pixel2style2pixel

---

CartoonGAN

Kaggle: rkuo2000/cartoongan

---

Image Inpainting

DeepFill V2

Kaggle: rkuo2000/deepfill-v2

Edge Connect

Kaggle: rkuo2000/edge-connect

---

Sefa

Kaggle: rkuo2000/genforce-sefa

---

Virtual Try On (VTON)

DeepFashion Try On

Kaggle: rkuo2000/deepfashion-tryon

---

NeRF

Tiny-NeRF

Kaggle: https://www.kaggle.com/rkuo2000/tiny-nerf

Nerfies

Colab:

• Process a video into a Nerfie dataset
• Train a Nerfie
• Render a Nerfie video

---

3D Avatar

Github: facebookresearch/pifuhd

---

Talking Head Image Animation

Kaggle: https://www.kaggle.com/rkuo2000/talking-head-image-animation

---

Music Seperation

Kaggle: rkuo2000/demucs
vocals

bass

drums

other

---

Voice Split

Kaggle: rkuo2000/voicesplit
input_mixed

predicted1

predicted2

---

SV2TTS

Github: CorentinJ/Real-Time-Voice-Cloning

This site was last updated December 22, 2022.