https://github.com/ben-louis/deep-image-analogy-pytorch

A python implementation of Deep-Image-Analogy based on pytorch.
https://github.com/ben-louis/deep-image-analogy-pytorch

pytorch style-transfer

Last synced: about 23 hours ago
JSON representation

A python implementation of Deep-Image-Analogy based on pytorch.

• Host: GitHub
• URL: https://github.com/ben-louis/deep-image-analogy-pytorch
• Owner: Ben-Louis
• License: mit
• Created: 2018-03-01T01:56:43.000Z (over 6 years ago)
• Default Branch: master
• Last Pushed: 2021-12-11T08:31:53.000Z (almost 3 years ago)
• Last Synced: 2024-11-07T21:47:07.507Z (9 days ago)
• Topics: pytorch, style-transfer
• Language: Python
• Homepage:
• Size: 25.2 MB
• Stars: 181
• Watchers: 7
• Forks: 42
• Open Issues: 4
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Deep-Image-Analogy

[![996.ICU](https://img.shields.io/badge/link-996.icu-red.svg)](https://996.icu)

This project is a python implementation of [Deep Image Analogy](https://arxiv.org/abs/1705.01088).

## Some results

![](results/results.jpg)

## Requirements

 - python3

 - opencv3

   If you use anaconda, you can install opencv3 by  ```conda install opencv```

 - [pytorch](http://pytorch.org/) (>= 1.7 for cpu-only mode, >= 1.3 for gpu mode)

 - [pycuda](https://pypi.org/project/pycuda/) (optional)

## Usage (demo)

#### Use GPU

```bash

python main.py --use_cuda

```

#### CPU only

```bash

python main.py

```

I train a series of invert blocks which estimate the inputs given the outputs for each stage of Vgg19. 

The results are slightly worse than optimization-based method.

## Acknowledgments

My project acknowledge the official code [Deep-Image-Analogy](https://github.com/msracver/Deep-Image-Analogy), [pytorch](http://pytorch.org/), and [another pytorch implementation](https://github.com/harveyslash/Deep-Image-Analogy-PyTorch). Especially, thanks for the authors of this amazing algorithm.