/poisson-image-editing

Python implementation of Poisson image editing

Primary LanguagePython

Poisson Image Editing

William Emmanuel

Computational Photography Project

Combines a source image, target image, and mask to create a content-aware blend.

Based on Patrick Perez, Michel Gangnet, and Andrew Blake's original paper on Poisson Image Editing http://www.cs.virginia.edu/~connelly/class/2014/comp_photo/proj2/poisson.pdf

source + mask + target = result

Usage

vagrant up && vagrant ssh

python main.py

Dependencies: numpy, scipy.sparse, opencv

Place a target image, source image, and mask image in a directory in /input. One directory is needed per image group to be processed. The result will be placed in the /output directory

NOTE: To process the 3 examples provided in /input, this program takes about 35 seconds.