Giraffe generator by rybka 🐟 plum

Giraffe generator made for Motorola Science Cup contest

Project was created on Linux where configuration shouldn't cause much problems.

Usage

Requirements

python3 is required to run run.py script. There shouldn't be a problem with main, but libpng may be required.

Running

Use run.py script if you want to enter parameters manually. You can also import the script and use a provided function to automate it.

The other way is to write directly to stdin of main.

Examples

There are two types of examples:

*.py - those are run with python3 {example-file}
*.txt - those are run with ./main < {example-file}

basic_input.txt
---
giraffe - name of the file
512 512 - width and height
1 - scale
0 - giraffe type (Reticulated giraffe)
0 - skip colors customization
0 - skip patches customization

colored_giraffe.txt
---
colored_giraffe.png - name of the file
512 512 - width and height
2.0 - scale
3 - giraffe type (Kordofan giraffe)
1 - customize colors
202 127 79 - patch color
232 223 202 - gap color
112 74 56 - middle color 
237 208 144 - secondary color
0 - skip patches customization

Format

Generator (main) reads parameters from stdin in the following format:

{general parameters}
0 | (1 {color parameters})
0 | (1 {patch parameters})

Parameter order

Note: All parameters are in relation to 512x512 image. Image size 'crops' and has no effect on the pattern

General

filename {string} - of result image
width {uint} - of result image
height {uint} - of result image
scale {float} - scales area of texture elements
giraffe index {uint} - number between 0 and 8 representing giraffe

Colors

patch {color} - overall, average color of a patch
gap {color} - space between the patches
middle {color} - color deeper inside the patch, usually darker
secondary {color} - adds some variation

Patches

Size

scale {float} - scale of patch area - greater scale = bigger patches and fewer of them
(min and max) gap {float} - how far apart the patches are
min to max area ratio {float} - measure of how different in size the patches are
standard deviation {float} - size is randomly selected with normal distribution

Edges - created by adding noise to generated patches. Each property has 3 values: frequency {float}, strength {float}, octaves {uint}