VinceMarron / Style_transfer
Style Transfer as Optimal Transport
Stars: ✭ 278
Labels
Projects that are alternatives of or similar to Style transfer
Toon Me
A Deep Learning project to Toon Portrait Images
Stars: ✭ 276 (-0.72%)
Mutual labels: jupyter-notebook
Demo Chinese Text Binary Classification With Bert
Stars: ✭ 276 (-0.72%)
Mutual labels: jupyter-notebook
Introduction Datascience Python Book
Introduction to Data Science: A Python Approach to Concepts, Techniques and Applications
Stars: ✭ 275 (-1.08%)
Mutual labels: jupyter-notebook
Data Science Learning
Repository of code and resources related to different data science and machine learning topics. For learning, practice and teaching purposes.
Stars: ✭ 273 (-1.8%)
Mutual labels: jupyter-notebook
Latest News Classifier
Master in Data Science Final Project
Stars: ✭ 276 (-0.72%)
Mutual labels: jupyter-notebook
Unrolled gan
Unrolled Generative Adversarial Networks
Stars: ✭ 277 (-0.36%)
Mutual labels: jupyter-notebook
Quietnet
Simple chat program that communicates using inaudible sounds
Stars: ✭ 2,924 (+951.8%)
Mutual labels: jupyter-notebook
Transformer
Implementation of Transformer model (originally from Attention is All You Need) applied to Time Series.
Stars: ✭ 273 (-1.8%)
Mutual labels: jupyter-notebook
Adaptnlp
An easy to use Natural Language Processing library and framework for predicting, training, fine-tuning, and serving up state-of-the-art NLP models.
Stars: ✭ 278 (+0%)
Mutual labels: jupyter-notebook
Generative Adversarial Networks
Tutorial on GANs
Stars: ✭ 275 (-1.08%)
Mutual labels: jupyter-notebook
Pgcn
Graph Convolutional Networks for Temporal Action Localization (ICCV2019)
Stars: ✭ 276 (-0.72%)
Mutual labels: jupyter-notebook
Trading Bot
Stock Trading Bot using Deep Q-Learning
Stars: ✭ 273 (-1.8%)
Mutual labels: jupyter-notebook
Generative models tutorial with demo
Generative Models Tutorial with Demo: Bayesian Classifier Sampling, Variational Auto Encoder (VAE), Generative Adversial Networks (GANs), Popular GANs Architectures, Auto-Regressive Models, Important Generative Model Papers, Courses, etc..
Stars: ✭ 276 (-0.72%)
Mutual labels: jupyter-notebook
Pyopenpose
Python bindings for the Openpose library
Stars: ✭ 277 (-0.36%)
Mutual labels: jupyter-notebook
Style Transfer as Optimal Transport
An algorithm that transfers the distribution of visual characteristics, or style, of a reference image onto a subject image via an Optimal Transport plan.
How it Works:
- For a general description and pictures: style-transfer-theory.pdf.
- For an explanation of the use of L2-Wasserstein distance instead of the loss function from the Gatys et. al framework see why_wasserstein.ipynb.
- Demo video
tl;dr - Passes a subject and style image through the convolution layers of the vgg19 network. Extracts probabalistic descriptions (first two moments) of the convolution filter activations. Calculates L2-Wasserstein distance between these probability distributions and then modifies the subject image optimally to minimize this distance.
Running
Requires:
- Python 3 (w/ NumPy and PIL)
- Tensorflow (tested w/ version 1.3)
- 'vgg_conv.npy' binary (115.5 MB) which contains calibrated convolution filters from vgg19 network. Can be downloaded here [md5sum: bf8a930fec201a0a2ade13d3f7274d0e]
Basic Usage from Command Line:
python basic_styletrans.py --subject media/wave_small.jpg --style media/kngwa_small.jpg --output media/wave_kngwa.jpg --vggnet vgg_conv.npy
Output:
loss: 118,500,040.00
ALL DONE
post clip | time: 49.0 final loss: 2,771,486.50
synthesized image saved: media/wave_kngwa.jpg
Note that the project description data, including the texts, logos, images, and/or trademarks,
for each open source project belongs to its rightful owner.
If you wish to add or remove any projects, please contact us at [email protected].