Approximates images in svg format. Now supports Bezier curves!

input image	output svg (polygonal)

Use the recursive submodule option to clone the necessary submodules.

git clone --recurse-submodules https://github.com/loicland/img2svg

Otherwise, clone the following repo in the root:

• https://gitlab.com/1a7r0ch3/parallel-cut-pursuit
• https://gitlab.com/1a7r0ch3/multilabel-potrace
• https://gitlab.com/1a7r0ch3/grid-graph

You need numpy, matplotlib

pip install numpy matplotlib

In a command prompt run:

python setup.py build_ext

Test with

python img2svg.py

python img2svg -f filename -r reg_strength -c contour_color -s stroke_width -o out_size -p output_path

With

• -f --file: path to image
• -r --reg_strength : how much simplification to do. Default = 1.0
• -lc --line_color: color of the contour, if any. Default = '' (no contour). Exemple : 'red'
• -lw --line_width: width of the contour (only if contour_color != ''). Default = 2.0 (thick)
• -o --out_size: size of the largest side of the image. Default = 500
• -p --output_path: output path. Default = replace the image extension by `svg`
• -a --apply: Function to apply before partition: sqrt, log, none (default)
• -s --smooth: Smoothing term. 0 = polygonal, >0 cubic Bezier curves. Default = 1.0
• -lt --line_tolerance: How far are lines allowed to deviate from the borders. Default = 1.0
• -ct --curve_tolerance: max difference area ratio diff between original and simplified polygons. Default=0.2