https://github.com/sepandhaghighi/zstyle
Z-Style : Generative Z-Style Avatars
https://github.com/sepandhaghighi/zstyle
Last synced: 14 days ago
JSON representation
Z-Style : Generative Z-Style Avatars
- Host: GitHub
- URL: https://github.com/sepandhaghighi/zstyle
- Owner: sepandhaghighi
- License: mit
- Created: 2021-10-22T08:37:25.000Z (about 4 years ago)
- Default Branch: master
- Last Pushed: 2021-10-22T10:10:54.000Z (about 4 years ago)
- Last Synced: 2025-04-01T04:52:14.503Z (7 months ago)
- Language: Python
- Homepage:
- Size: 29.6 MB
- Stars: 5
- Watchers: 2
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
Z-Style
----------
## Overview
In Z-Style I tried to generate generative arts for [ZILLIQA ON GITCOIN: NFTs BEYOND ART] Generative Art Pieces that can serve as unique memorabilia or even avatars.
This project is based on our other library called Samila and consist of two main parts:
1. Generative Wallpapers
2. Generative Z-Style Avatars
## Requirements
1. [Samila](https://github.com/sepandhaghighi/samila)
2. [Art](https://github.com/sepandhaghighi/art)
Run `pip3 install -r requirements.txt`
## Generative Wallpapers
- Run `python3 rand_gen.py`
This part lets you create infinite unique wallpapers based on many thousand points. The position of every single point is calculated by a formula, which has random parameters. Because of the random numbers, every image looks different.
Parameters :
1. Input formula
2. Projection
3. Color
4. Spot size
### Sample1
```
F1 : 30.586*y**2*x**1.59-27.055*abs(y-x)**1.646-22.722*math.cos(y**2)**1.108-26.326*math.ceil(x*y)**1.326+35.621*math.ceil(x-y)**1.706+16.248*abs(x**2*y)**1.636
F2 : 45.519*math.sin(y)**1.374+2.565*abs(y**2*x)**1.416+19.316*x**1.703-14.939*x**2**1.78+31.781*math.floor(y**2*x)**1.445
Seed : 690765
Color : (0.167, 0.2, 0.184)
Projection : Projection.RECTILINEAR
Spot Size : 0.29045928118447506
```
### Sample2
```
F1 : 46.739*math.floor(y**2*x)**1.444-18.103*math.sin(y**2)**1.871-11.165*math.cos(x**2)**1.07-14.044*math.cos(y**2*x)**1.885-23.404*math.sin(x)**1.175+42.816*math.sin(y-x)**1.572
F2 : 37.384*abs(x)**1.811+33.06*math.sin(x*y)**1.415+16.468*abs(x)**1.817+14.806*math.ceil(y-x)**1.042+45.238*abs(y)**1.204
Seed : 202731
Color : (0.465, 0.35, 0.247)
Projection : Projection.RECTILINEAR
Spot Size : 1.9499281137566082
```
### Sample3
```
F1 : 11.516*x+y**1.034-42.219*math.sin(x)**1.423
F2 : 41.56*math.sin(y**2)**1.011
Seed : 728376
Color : (0.206, 0.164, 0.005)
Projection : Projection.POLAR
Spot Size : 1.7475050745556027
```
For more example take a look at [Wallpapers](Wallpapers/) folder
Try online : [](https://mybinder.org/v2/gh/sepandhaghighi/zstyle/master)
* Run `rand_gen.ipynb`
## Z-Style Avatars
- Run `python3 z_style_gen.py`
This part lets you create infinite unique z-style avatars based on many thousand points. The position of every single point is calculated by a formula, which has random parameters. Because of the random numbers, every image looks different.
Parameters :
1. Input formula
2. Color
3. Spot size
### Sample1
```
F1 : F2
F2 : F3
Seed : 56985
Color : (0.465, 0.181, 0.148)
Spot Size : 1.6153854689146487
```
### Sample2
```
F1 : F1
F2 : F2
Seed : 420135
Color : (0.28, 0.486, 0.087)
Spot Size : 1.4430856043963218
```
### Sample3
```
F1 : F2
F2 : F3
Seed : 244129
Color : (0.233, 0.354, 0.119)
Spot Size : 0.6440937532451008
```
For more example take a look at [Z_Style](Z_Style/) folder
Try online : [](https://mybinder.org/v2/gh/sepandhaghighi/zstyle/master)
* Run `z_style_gen.ipynb`
## Technical Details
A transformation between a square-shaped space from the Cartesian coordinate system to any arbitrary coordination like [Polar coordinate system](https://en.wikipedia.org/wiki/Polar_coordinate_system).
### Example
We have set of points in the first space (left square) which can be define as follow:
And bellow functions are used for transformation:
```pycon
>>> def f1(x,y):
result = random.uniform(-1,1) * x**2 - math.sin(y**2) + abs(y-x)
return result
>>> def f2(x,y):
result = random.uniform(-1,1) * y**3 - math.cos(x**2) + 2*x
return result
```
here we uses `Projection.POLAR` so later space will be the polar space and we have:
```pycon
>>> g = GenerativeImage(f1,f2)
>>> g.generate(seed=10)
>>> g.plot(projection=Projection.POLAR)
```
