Overview
Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only!
Explore the first part of a comprehensive three-part tutorial on one-shot face stylization using JoJoGAN. Learn how to transfer human-centric face landmarks and apply extreme style references with minimal data and training time. Dive into JoJoGAN's introduction, complete a Google Colab exercise, and discover how to use existing style models for inference. Master the process of training your own style using single images, multiple images, and even animal image sources. Gain hands-on experience with StyleGAN and JoJoGAN for face style transfer, setting the foundation for advanced techniques covered in subsequent parts of the series.
Syllabus
- Video Start
- Why JoJoGAN
- JoJo GAN Resources
- Video Content Intro
- Google Colab Setup
- Loading Python Modules
- Loading Source Image
- Selecting Existing Style
- You own Face Style
- Fine-tuning GAN
- Checking Results
- Custom Style with Animal Face
- GitHub Resources
Taught by
Prodramp