Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/vvmnnnkv/nodejs-style-transfer
Sample Node.js app for image style transfer using libtorchjs
https://github.com/vvmnnnkv/nodejs-style-transfer
libtorch libtorchjs nodejs pytorch style-transfer
Last synced: 23 days ago
JSON representation
Sample Node.js app for image style transfer using libtorchjs
- Host: GitHub
- URL: https://github.com/vvmnnnkv/nodejs-style-transfer
- Owner: vvmnnnkv
- Created: 2019-01-07T14:24:30.000Z (about 6 years ago)
- Default Branch: master
- Last Pushed: 2022-12-10T19:23:55.000Z (about 2 years ago)
- Last Synced: 2024-02-24T02:31:17.269Z (12 months ago)
- Topics: libtorch, libtorchjs, nodejs, pytorch, style-transfer
- Language: Python
- Homepage: https://nodejs-style-transfer.herokuapp.com
- Size: 6.11 MB
- Stars: 20
- Watchers: 1
- Forks: 5
- Open Issues: 4
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Image Style Transfer with Node.JS
Demo of image style transfer app based on Node.js and Express
that uses [fast neural style transfer](https://github.com/pytorch/examples/tree/master/fast_neural_style) model created and traced with Pytorch.
## Usage
Open the site on your mobile or desktop, shoot or upload something, stylize :-)
Live demo: https://nodejs-style-transfer.herokuapp.com
## Why?
This app is inspired by Lessons 6 & 9 of [Facebook Pytorch Scholarship](https://www.udacity.com/facebook-pytorch-scholarship).
It shows how easy it is possible to deploy high-performance ML models these days combining
such popular platforms as Node.js and NPM with newly released Torch Script feature of Pytorch.
Apart from that, Node.js non-blocking I/O seems to be good fit for running NN inference in Worker Pool.
The heavy lifting of image processing is done by [libtorch C++ library](https://pytorch.org/cppdocs/), which is wrapped in Node.js module [libtorchjs](https://www.npmjs.com/package/libtorchjs)
created specially for this app to expose some bits of libtorch in JS.
## Running
Currently supported platforms are Linux and Windows. libtorch does have Mac variant,
but it's not included in [libtorchjs](https://www.npmjs.com/package/libtorchjs) yet.
GPU is also not supported yet.
### Linux or Windows
Checkout repo, install npm libraries, start app:
```
$ npm i
$ npm run start
```
Open http://localhost:3000/
### Docker
Checkout repo, build image and run:
```
$ docker build --tag=nodejs-style-transfer .
$ docker run -P nodejs-style-transfer
```
Open http://localhost:3000/
## How to Add Style
This app has just one style included.
The easiest way to create more models is to install [`fast_neural_style`](https://github.com/pytorch/examples/tree/master/fast_neural_style) from Pytorch examples
and add following code in the end of `stylyze` function in
[fast_neural_style/neural_style/neural_style.py](https://github.com/pytorch/examples/blob/master/fast_neural_style/neural_style/neural_style.py#L122):
```python
traced_script_module = torch.jit.trace(style_model, content_image)
traced_script_module.save("style_model.pt")
```
And after that, execute stylization script on any image and this will write out `style_model.pt`.
Note that traced model will keep the size of your image and will be able to work only with same dimensions.
For this demo app, image size is fairly small (`224x224`) so it's fast even on CPU.
See [libtorch tutorial](https://pytorch.org/tutorials/advanced/cpp_export.html) for more info about model tracing.
## Acknowledgements
* Inspiration: https://www.udacity.com/facebook-pytorch-scholarship
* Model: https://github.com/pytorch/examples/tree/master/fast_neural_style
* Model tracing tutorial: https://pytorch.org/tutorials/advanced/cpp_export.html
* Canvas & EXIF helpers: https://stackoverflow.com/a/40867559 https://stackoverflow.com/a/21961894
* Spinners: https://loading.io/spinner/fidget-spinner
## Disclamer
Note that it's just a toy project and [libtorchjs](https://www.npmjs.com/package/libtorchjs) is too immature to use in production.