An image classifier built around the desire to match two images, one a real life portrait and the other an artificial manga drawing. shown here:
It runs different similarity checks such as:
- Structural similarity indexes
- MSE
- SSIM
- Edge detection and tracer into black & white
- Laplace Transforms
- Shape and object tracing
- Contours
These instructions will get you a copy of the project up and running on your local machine for development and testing purposes.
- Two images, required to be the exact same dimensions.
What things you need to install the software and how to install them
-
Python3 is required, please upgrade from Python2 if you have not already done so.
-
Make sure to have the following modules installed:
skimage
andpython-opencv
.
- Drop in your own images to the
./images/
directory and run the classifier:
python3 compare.py
- A matlab plotter GUI will popup indicating the similarity factor between the two input images.
- Neural networking infrastructure to create layers to classify the edges (similarity)
- https://docs.opencv.org/3.2.0/d5/d69/tutorial_py_non_local_means.html
- http://www.robindavid.fr/opencv-tutorial/chapter5-line-edge-and-contours-detection.html
- https://people.eecs.berkeley.edu/~aramdas/reports/DBreport.pdf
- https://www.ml.uni-saarland.de/code/GraphDemo/HeinLuxburg_SlidesSimilarityGraphs.pdf
- http://www.cs.cornell.edu/~dph/papers/HKR-TPAMI-93.pdf