Adversarial Framework for (non-) Parametric Image Stylisation Mosaics

Overview

Fully Adversarial Mosaics (FAMOS)

Pytorch implementation of the paper "Copy the Old or Paint Anew? An Adversarial Framework for (non-) Parametric Image Stylization" available at http://arxiv.org/abs/1811.09236.

This code allows to generate image stylisation using an adversarial approach combining parametric and non-parametric elements. Tested to work on Ubuntu 16.04, Pytorch 0.4, Python 3.6. Nvidia GPU p100. It is recommended to have a GPU with 12, 16GB, or more of VRAM.

Parameters

Our method has many possible settings. You can specify them with command-line parameters. The options parser that defines these parameters is in the config.py file and the options are parsed there. You are free to explore them and discover the functionality of FAMOS, which can cover a very broad range of image stylization settings.

There are 5 groups of parameter types:

  • data path and loading parameters
  • neural network parameters
  • regularization and loss criteria weighting parameters
  • optimization parameters
  • parameters of the stochastic noise -- see PSGAN

Update Febr. 2019: video frame-by-frame rendering supported

mosaicGAN.py can now render a whole folder of test images with the trained model. Example videos: lion video with Münich and Berlin

Just specify

python mosaicGAN.py --texturePath=samples/milano/ --contentPath=myFolder/ --testImage=myFolder/ 

with your myFolder and all images from that folder will be rendered by the generator of the GAN. Best to use the same test folder as content folder for training. To use in a video editing pipeline, save all video frames as images with a tool like AVIDEMUX, train FAMOS and save rendered frames, assemble again as video. Note: this my take some time to render thousands of images, you can edit in the code VIDEO_SAVE_FREQ to render the test image folder less frequently.

Update Jan. 2019: new functionality for texture synthesis

Due to interest in a new Pytorch implementation of our last paper "Texture Synthesis with Spatial Generative Adversarial Networks" (PSGAN) we added a script reimplementing it in the current repository. It shares many components with the texture mosaic stylization approach. A difference: PSGAN has no content image and loss, the generator is conditioned only on noise. Example call for texture synthesis:

python PSGAN.py --texturePath=samples/milano/ --ngf=120 --zLoc=50 --ndf=120 --nDep=5 --nDepD=5 --batchSize=16

In general, texture synthesis is much faster than the other methods in this repository, so feel free to add more channels and increase th batchsize. For more details and inspiration how to play with texture synthesis see our old repository with Lasagne code for PSGAN.

Usage: parametric convolutional adversarial mosaic

We provide scripts that have a main loop in which we (i) train an adversarial stylization model and (ii) save images (inference mode). If you need it, you can easily modify the code to save a trained model and load it later to do inference on many other images, see comments at the end of mosaicGAN.py.

In the simplest case, let us start an adversarial mosaic using convolutional networks. All you need is to specify the texture and content folders:

python mosaicGAN.py --texturePath=samples/milano/ --contentPath=samples/archimboldo/

This repository includes sample style files (4 satellite views of Milano, from Google Maps) and a portrait of Archimboldo (from the Google Art Project). Our GAN method will start running and training, occasionally saving results in "results/milano/archimboldo/" and printing the loss values to the terminal. Note that we use the first image found in contentPath as the default full-size output image stylization from FAMOS. You can also specify another image file name testImage to do out-of-sample stylization (inference).

This version uses DCGAN by default, which works nicely for the convolutional GAN we have here. Add the parameter LS for a least squares loss, which also works nicely. Interestingly, WGAN-GP is poorer for our model, which we did not investigate in detail.

If you want to tune the optimisation and model, you can adjust the layers and channels of the Generator and Discriminator, and also choose imageSize and batchSize. All this will effect the speed and performance of the model. You can also tweak the correspondance map cLoss and the content loss weighting fContent

python mosaicGAN.py --texturePath=samples/milano/ --contentPath=samples/archimboldo/ --imageSize=192 --batchSize=8 --ngf=80 --ndf=80  --nDepD=5  --nDep=4 --cLoss=101 --fContent=.6

Other interesting options are skipConnections and Ubottleneck. By disabling the skip connections of the Unet and defining a smaller bottleneck we can reduce the effect of the content image and emphasize more the texture style of the output.

Usage: the full FAMOS approach with parametric and non-parametric aspects

Our method has the property of being able to copy pixels from template images together with the convolutional generation of the previous section.

python mosaicFAMOS.py  --texturePath=samples/milano/ --contentPath=samples/archimboldo/ --N=80 --mirror=True --dIter=2 --WGAN=True

Here we specify N=80 memory templates to copy from. In addition, we use mirror augmentation to get nice kaleidoscope-like effects in the template (and texture distribution). We use the WGAN GAN criterium, which works better for the combined parametric/non-parametric case (experimenting with the usage of DCGAN and WGAN depending on the architecture is advised). We set to use dIter=2 D steps for each G step.

The code also supports a slightly more complicated implementation than the one described in the paper. By setting multiScale=True a mixed template of images I_M on multiple levels of the Unet is used. In addition, by setting nBlocks=2 we will add residual layers to the decoder of the Unet, for a model with even higher capacity. Finally, you can also set refine=True and add a second Unet to refine the results of the first one. Of course, all these additional layers come at a computational cost -- selecting the layer depth, channel width, and the use of all these additional modules is a matter of trade-off and experimenting.

python mosaicFAMOS.py  --texturePath=samples/milano/ --contentPath=samples/archimboldo/ --N=80 --mirror=True --multiScale=True --nBlocks=1 --dIter=2 --WGAN=True

The method will save mosaics occasionally, and optionally you can specify a testImage (size smaller than the initial content image) to check out-of-sample performance. You can check the patches image saved regularly how the patch based training proceeds. The files has a column per batch-instance, and 6 rows showing the quantities from the paper:

  • I_C content patch
  • I_M mixed template patch on highest scale
  • I_G parametric generation component
  • I blended patch
  • \alpha blending mask
  • A mixing matrix

License

Please make sure to cite/acknowledge our paper, if you use any of the contained code in your own projects or publication.

The MIT License (MIT)

Copyright © 2018 Zalando SE, https://tech.zalando.com

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Owner
Zalando Research
Repositories of the research branch of Zalando SE
Zalando Research
🌊 River is a Python library for online machine learning.

River is a Python library for online machine learning. It is the result of a merger between creme and scikit-multiflow. River's ambition is to be the go-to library for doing machine learning on strea

OnlineML 4k Jan 03, 2023
A data preprocessing and feature engineering script for a machine learning pipeline is prepared.

FEATURE ENGINEERING Business Problem: A data preprocessing and feature engineering script for a machine learning pipeline needs to be prepared. It is

Pinar Oner 7 Dec 18, 2021
This jupyter notebook project was completed by me and my friend using the dataset from Kaggle

ARM This jupyter notebook project was completed by me and my friend using the dataset from Kaggle. The world Happiness 2017, which ranks 155 countries

1 Jan 23, 2022
Adversarial Framework for (non-) Parametric Image Stylisation Mosaics

Fully Adversarial Mosaics (FAMOS) Pytorch implementation of the paper "Copy the Old or Paint Anew? An Adversarial Framework for (non-) Parametric Imag

Zalando Research 120 Dec 24, 2022
Machine learning that just works, for effortless production applications

Machine learning that just works, for effortless production applications

Elisha Yadgaran 16 Sep 02, 2022
李航《统计学习方法》复现

本项目复现李航《统计学习方法》每一章节的算法 特点: 笔记摘要:在每个文件开头都会有一些核心的摘要 pythonic:这里会用尽可能规范的方式来实现,包括编程风格几乎严格按照PEP8 循序渐进:前期的算法会更list的方式来做计算,可读性比较强,后期几乎完全为numpy.array的计算,并且辅助详

58 Oct 22, 2021
Regularization and Feature Selection in Least Squares Temporal Difference Learning

Regularization and Feature Selection in Least Squares Temporal Difference Learning Description This is Python implementations of Least Angle Regressio

Mina Parham 0 Jan 18, 2022
A toolkit for making real world machine learning and data analysis applications in C++

dlib C++ library Dlib is a modern C++ toolkit containing machine learning algorithms and tools for creating complex software in C++ to solve real worl

Davis E. King 11.6k Jan 02, 2023
AutoTabular automates machine learning tasks enabling you to easily achieve strong predictive performance in your applications.

AutoTabular automates machine learning tasks enabling you to easily achieve strong predictive performance in your applications. With just a few lines of code, you can train and deploy high-accuracy m

Robin 55 Dec 27, 2022
Add built-in support for quaternions to numpy

Quaternions in numpy This Python module adds a quaternion dtype to NumPy. The code was originally based on code by Martin Ling (which he wrote with he

Mike Boyle 531 Dec 28, 2022
NCVX (NonConVeX): A User-Friendly and Scalable Package for Nonconvex Optimization in Machine Learning.

NCVX (NonConVeX): A User-Friendly and Scalable Package for Nonconvex Optimization in Machine Learning.

SUN Group @ UMN 28 Aug 03, 2022
Backprop makes it simple to use, finetune, and deploy state-of-the-art ML models.

Backprop makes it simple to use, finetune, and deploy state-of-the-art ML models. Solve a variety of tasks with pre-trained models or finetune them in

Backprop 227 Dec 10, 2022
It is a forest of random projection trees

rpforest rpforest is a Python library for approximate nearest neighbours search: finding points in a high-dimensional space that are close to a given

Lyst 211 Dec 29, 2022
Simple and flexible ML workflow engine.

This is a simple and flexible ML workflow engine. It helps to orchestrate events across a set of microservices and create executable flow to handle requests. Engine is designed to be configurable wit

Katana ML 295 Jan 06, 2023
A Lightweight Hyperparameter Optimization Tool 🚀

The mle-hyperopt package provides a simple and intuitive API for hyperparameter optimization of your Machine Learning Experiment (MLE) pipeline.

Robert Lange 137 Dec 02, 2022
Uplift modeling and causal inference with machine learning algorithms

Disclaimer This project is stable and being incubated for long-term support. It may contain new experimental code, for which APIs are subject to chang

Uber Open Source 3.7k Jan 07, 2023
CorrProxies - Optimizing Machine Learning Inference Queries with Correlative Proxy Models

CorrProxies - Optimizing Machine Learning Inference Queries with Correlative Proxy Models

ZhihuiYangCS 8 Jun 07, 2022
To-Be is a machine learning challenge on CodaLab Platform about Mortality Prediction

To-Be is a machine learning challenge on CodaLab Platform about Mortality Prediction. The challenge aims to adress the problems of medical imbalanced data classification.

Marwan Mashra 1 Jan 31, 2022
Client - 🔥 A tool for visualizing and tracking your machine learning experiments

Weights and Biases Use W&B to build better models faster. Track and visualize all the pieces of your machine learning pipeline, from datasets to produ

Weights & Biases 5.2k Jan 03, 2023