All Projects → tjwei → Quick Neural Art Transfer

tjwei / Quick Neural Art Transfer

Licence: mit
Theano/Lasagne based Neural artistic style transfer with Kivy GUI

Projects that are alternatives of or similar to Quick Neural Art Transfer

Icml2020
Notebook for comprehensive analysis of authors, organizations, and countries of ICML 2020 papers.
Stars: ✭ 42 (+0%)
Mutual labels:  jupyter-notebook
Tensorflow Programs And Tutorials
Implementations of CNNs, RNNs, GANs, etc
Stars: ✭ 1,012 (+2309.52%)
Mutual labels:  jupyter-notebook
Adaptive Multispeaker Separation
Adaptive and Focusing Neural Layers for Multi-Speaker Separation Problem
Stars: ✭ 42 (+0%)
Mutual labels:  jupyter-notebook
Infectious disease modelling
Stars: ✭ 42 (+0%)
Mutual labels:  jupyter-notebook
Bayesgan
Tensorflow code for the Bayesian GAN (https://arxiv.org/abs/1705.09558) (NIPS 2017)
Stars: ✭ 1,011 (+2307.14%)
Mutual labels:  jupyter-notebook
Computervision Recipes
Best Practices, code samples, and documentation for Computer Vision.
Stars: ✭ 8,214 (+19457.14%)
Mutual labels:  jupyter-notebook
Slow disentanglement
Towards Nonlinear Disentanglement in Natural Data with Temporal Sparse Coding
Stars: ✭ 41 (-2.38%)
Mutual labels:  jupyter-notebook
Ml Tutorial Notebooks
This depot contain tutorials for real beginners who want to understand machine learning by reading some code.
Stars: ✭ 42 (+0%)
Mutual labels:  jupyter-notebook
Tensorflow Seq2seq Tutorials
Dynamic seq2seq in TensorFlow, step by step
Stars: ✭ 1,011 (+2307.14%)
Mutual labels:  jupyter-notebook
Dl Twitch Series
Notebook from the Deep Learning Twitch Series on AWS (https://twitch.tv/aws)
Stars: ✭ 42 (+0%)
Mutual labels:  jupyter-notebook
Machine Learning From Scratch
Succinct Machine Learning algorithm implementations from scratch in Python, solving real-world problems (Notebooks and Book). Examples of Logistic Regression, Linear Regression, Decision Trees, K-means clustering, Sentiment Analysis, Recommender Systems, Neural Networks and Reinforcement Learning.
Stars: ✭ 42 (+0%)
Mutual labels:  jupyter-notebook
Keras Pytorch Avp Transfer Learning
We pit Keras and PyTorch against each other, showing their strengths and weaknesses in action. We present a real problem, a matter of life-and-death: distinguishing Aliens from Predators!
Stars: ✭ 42 (+0%)
Mutual labels:  jupyter-notebook
Pyhowfun
🌻 提供一系列的 Python 教學,讓初學者或是有程式基礎的人,快速融入 Python 的世界
Stars: ✭ 42 (+0%)
Mutual labels:  jupyter-notebook
Mp
公众号《林骥》的文章和代码
Stars: ✭ 41 (-2.38%)
Mutual labels:  jupyter-notebook
2016 Pydata Carolinas Pandas
Material for Pandas Tutorial at Pydata Carolinas 2016
Stars: ✭ 42 (+0%)
Mutual labels:  jupyter-notebook
Siren Fastai2
Unofficial implementation of 'Implicit Neural Representations with Periodic Activation Functions'
Stars: ✭ 42 (+0%)
Mutual labels:  jupyter-notebook
Ntm Tensorflow
"Neural Turing Machine" in Tensorflow
Stars: ✭ 1,013 (+2311.9%)
Mutual labels:  jupyter-notebook
Spark Sklearn Airbnb Predict
Code example to predict prices of Airbnb vacation rentals, using scikit-learn on Spark with spark-sklearn, on MapR.
Stars: ✭ 42 (+0%)
Mutual labels:  jupyter-notebook
Pstocky
股票小数据
Stars: ✭ 42 (+0%)
Mutual labels:  jupyter-notebook
Notebooks
Sample Notebooks for PipelineAI
Stars: ✭ 42 (+0%)
Mutual labels:  jupyter-notebook

Quick-Neural-Art-Transfer

Theano/Lasagne based Neural artistic style transfer with Kivy GUI

based on https://github.com/Lasagne/Recipes/tree/master/examples/styletransfer an implementation of the algorithm described in "A Neural Algorithm of Artistic Style" (http://arxiv.org/abs/1508.06576) by Gatys, Ecker and Bethge.

use VGG19 model http://www.robots.ox.ac.uk/~vgg/research/very_deep/ ( wget https://s3.amazonaws.com/lasagne/recipes/pretrained/imagenet/vgg19_normalized.pkl to get the pickled file)

Several modification are made to speed up the process.

  1. Doing the transfer in pyramid manner to speed up and seems generating better results.
  2. Has an option to use ADAM as the optimizer.
  3. Use shared memory and can precompile theanot functions.
  4. Has a server and a Kivy GUI client, so you can capture image using a mobile device and the use the GPU on a remote machine. Usage:
import neuralStyle as ns
import cv2
ns.precompute() # compile theano functions
for x in ns.p_transfer("photo.jpg", "art.jpg"):
  cv2.imshow("output", x)

or you can use nstest.py photo.jpg art1.jpg art2.jpg art3.jpg to test it.

Demo program

The demo program with kivy UI is designed for an exhibition, which takes user's input from webcam and rendered it with a style.

The program expect an gpu of gtx 980 or better. To run this program, you will need a gpu at least as powerful as gtx 965m. If your gpu is slower, than you will need to chang zmq.RCVTIMEO to a larger value(more than 15 seconds) First run the server,

python nserver.py 

you will need to wait a while it for theano function to be compiled, it may take up to 2 minutes. Then you can run the kivy client. it requires a webcam and cv2 installed. Then you can run the client

python main.py

Currently, all messages are in Chinese.

content and styles Many image files are from https://github.com/jcjohnson/neural-style and some are public domain

CC-By http://agf81.deviantart.com/art/Ivy-Texture-7-326197006?q=gallery%3ACreative-Commons%2F26171837&qo=33 Ivy texture

CC-BY-SA http://maadmann.deviantart.com/art/Fullcolor-437966862?q=gallery%3ACreative-Commons%2F26171826&qo=88 fullcolor

CC-BY https://www.flickr.com/photos/fontplaydotcom/24795772945/in/[email protected]/ color squares

CCB-BY-SA http://oilsoaked.deviantart.com/art/holiday-188505340?q=gallery%3ACreative-Commons%2F26171826&qo=373 Holidy

CC-BY https://www.flickr.com/photos/two/236883835/in/photolist-mW6j4-mW6kn-Jy6XP-Jy9fR-JxQD8-mW6gp-JxQGc-Jy5Xf-Jy3ES-mW6fr-Jy9eM-mW6hW-JxQqD-JxMdm-mW6nF-JxQtD-JxQxR-mW6mA-JxLZS-Jy72p-JxM2N-Jy3Hh-JxLU7-JxLWb-JxQsz-oH3ixB-4Nri37-JxQJr-JxQFe-Jy3AY-JxMcq-Jy3xG-JxQvT-Jy95i-Jy99n-JxQz2-a97MmE-JxQm6-Jy5V5-a97Nqb-a97PVu-Jy97c-a94ZyK-ADmwE6-B3g6KV-ADmtTV-BysQ61-a97Ptd-a94Zex-dwVdG ndhu

CC-BY https://www.flickr.com/photos/jphotos/5998521863/in/photolist-mW6gp-JxQGc-Jy5Xf-Jy3ES-mW6fr-Jy9eM-mW6hW-JxQqD-JxMdm-mW6nF-JxQtD-JxQxR-mW6mA-JxLZS-Jy72p-JxM2N-Jy3Hh-JxLU7-JxLWb-JxQsz-oH3ixB-4Nri37-JxQJr-JxQFe-Jy3AY-JxMcq-Jy3xG-JxQvT-Jy95i-Jy99n-JxQz2-a97MmE-JxQm6-Jy5V5-a97Nqb-a97PVu-Jy97c-a94ZyK-ADmwE6-B3g6KV-ADmtTV-BysQ61-a97Ptd-a94Zex-dwVdG-fJCm8H-ih4o2w-farbgP-ez8gpV-a244iL ndhu2

CC-BY https://www.flickr.com/photos/kurtbudiarto/7257851556/in/photolist-c4mopE-jZZkdx-8bohuZ-bjLvVe-c4wRru-gkp48t-hPNbbL-qfySJz-c84E3h-fBV6ve-aijVu7-fCaHPs-qZ7XnX-f8zn77-eQRUQ5-695MbG-dQNbTm-rowN34-6o4FFz-d3MjtL-9htwN2-r6Yib1-fCau8q-fCaCh9-551xgk-bkueEX-fBVtqt-pbJvac-fCbdY9-aRQ2oH-oUai6n-fCbaRs-fKzB9m-fCbmqA-7kBUGU-eX5Ce9-fBVu9x-ftqE1w-aihyDg-bku5n4-54rawt-nixm7H-dDjykx-55JGBX-c4mnWy-bkudkH-pRMC8L-55JGWa-5A9m5g-6uijiE/ smile face

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].