Style Transfer Part 1: Training a model with on Spell with Yining Shi

Channel:

The Coding Train

Subscribers:

1,740,000

Published on October 1, 2018 1:01:15 PM ● Video Link: https://www.youtube.com/watch?v=STHRNIJc-vI

Duration: 30:59

17,417 views

325

In this video, Yining Shi demonstrates how to train a "Style Transfer Model" using Spell (Sign up here: https://spell.run/codingtrain).
In Part 2, Yining will use this trained model to style a real-time image, in browser, using ml5.js and p5.js.
#ThisDotStyle #StyleTransfer #MachineLearning

🎥 Part 2: https://youtu.be/S_I0SGAO73A

This video is sponsored by Spell.

🔗 Style Transfer Example: https://yining1023.github.io/styleTransfer_spell/
🔗 Detailed Instructions: https://github.com/yining1023/styleTransfer_spell/
🔗 ml5.js: https://ml5js.org/
🔗 p5.js: https://p5js.org
🔗 Fast Style Transfer in TensorFlow by Logan Engstrom: https://github.com/lengstrom/fast-style-transfer/
🔗 Machine Learning 101: https://spell.run/docs/core_concepts/#machine-learning-101
🔗 What Neural Networks See by Gene Kogan: https://experiments.withgoogle.com/what-neural-nets-see
🔗 Fast style transfer in deeplearn.js by Reiichiro Nakano: https://github.com/reiinakano/fast-style-transfer-deeplearnjs

🎥 To learn more about style transfers: https://youtu.be/gye9hSIrRWI
🎥 Brick Breaker with Yining Shi: https://youtu.be/5kEPixL8JoU
🎥 Intro to Spell: https://youtu.be/ggBOAPtFjYU
🎥 Python and Virtualenv: https://youtu.be/nnhjvHYRsmM

Work Cited:
📙 A Neural Algorithm of Artistic Style [Leon A. Gatys, Alexander S. Ecker, Matthias Bethge]
📙 Perceptual Losses for Real-Time Style Transfer and Super-Resolution [Justin Johnson, Alexandre Alahi, Li Fei-Fei]
📙 Artistic style transfer for videos [Manuel Ruder, Alexey Dosovitskiy, Thomas Brox]
📙 Deep Photo Style Transfer [Fujun Luan, Sylvain Paris, Eli Shechtman, Kavita Bala]
📙 Visual Attribute Transfer through Deep Image Analogy [Jing Liao, Yuan Yao, Lu Yuan, Gang Hua, Sing Bing Kang]
📙 Universal Style Transfer via Feature Transforms [Yijun Li, Chen Fang, Jimei Yang, Zhaowen Wang, Xin Lu, Ming-Hsuan Yang]

Yining Shi is an artist and researcher who is interested in building tools to craft a better learning experience for people. She is also an adjunct professor at Interactive Telecommunications Program (ITP) at NYU, where she teaches Machine Learning for the Web class. She also contributes to various open source projects like p5.js, ml5.js. She currently works at Sourcemap as a Senior Software Engineer.

Yining Shi's website: http://1023.io/

📄 Code of Conduct: https://github.com/CodingTrain/Code-of-Conduct