Here is my python source code for Photomosaic generator. To the best of my knowledge, my code is the shortest one you could find out for this goal. With my code, you could:
- Given input image, you could generate Photomosaic art stored under image formats (.png, .jpg, ...)
- Given input video, you could generate Photomosaic art stored under video formats (.avi, .mp4, ...)
In order to run my script, you first need to prepare a set of photos put in the same folder (my default folder is image_pool). These photos could be in any image format (jpg, png, ...) with any resolution. In this repository, I put all photos used to generate the demo for video2video case (Taylor Swift's photos taken from the internet). For image2image case, I used photos from VOC2012 dataset link.
By running the sript video2video.py, we will have following output:
Output video
By running the sript image2image.py, we will have following output:
Output image
Input image
- python 3.6
- cv2
- numpy