Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/shvbsle/sputils
Utils and Notebooks for building deep-learing model with superpaper
https://github.com/shvbsle/sputils
dataset deep-learning generative-art sketch-rnn
Last synced: about 14 hours ago
JSON representation
Utils and Notebooks for building deep-learing model with superpaper
- Host: GitHub
- URL: https://github.com/shvbsle/sputils
- Owner: shvbsle
- Created: 2020-06-14T08:02:37.000Z (over 4 years ago)
- Default Branch: master
- Last Pushed: 2020-06-14T12:37:45.000Z (over 4 years ago)
- Last Synced: 2024-11-09T01:58:49.295Z (about 2 months ago)
- Topics: dataset, deep-learning, generative-art, sketch-rnn
- Language: Jupyter Notebook
- Homepage: https://superpaper.netlify.app/
- Size: 743 KB
- Stars: 0
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Superpaper Utils
Utils and Notebooks for building deep-learing models with data-generated from [superpaper web-app](https://superpaper.netlify.app/).P.S: Most images in dataset are gonna be low in quality as thats how good my art-skills are
# Data
The stroke-data is exported as a json file. Use the ExploreStrokes.ipynb notebook to use utils for exploring data.
Currently, a single data-point comprises on 3 elements:
1) Stroke JSON
2) Base image
3) Sketch Layer---
## 1. Stroke JSON
```
{
"description": "",
"stroke": [{
"layer": "",
"type": "",
"memento": [
[, , , ],
[809, 124, -310, -10],
.
.
[]
]
}, {
"layer": "Layer 1",
"type": "mouseup",
"memento": [
[811, 126, -310, -10],
[811, 126, -310, -10],
.
.
.
[]
]
}
.
.
.],
"device_type": "pc",
"canvas_h": 649,
"canvas_w": 999
}
```Note that an empty entry in memento (i.e. []) signifies end of stroke
## 2. Base Image
This is the image used as reference for drawing the data![alt text]()
## 3. Sketch Layer
This is the image that was exported after drawing (p.s forgive me for this garbage trace)![alt text]()
---
# TODO
- [ ] Add Sketch-RNN wrapper
- [X] Create section for comic