Restoring distorted images (distortions can be translation, rotation and scaling) and non-ML recognition of 8 different shapes.
You are given 2 grayscale images, image before distortion and same image after distortion. Find the (x, y) coordinates of the centroid (center of mass) of every shape in the 2 images you are given. Classify each shape into 6 categories (cross, spiral, circle, flower, star, donut). Given the (x, y) positions of the stars in the undistorted image, find their (x, y) positions in the distorted image. Distortions can be: translation, rotation, scaling or their combination.
Source and distorted images can be found in PublicDataSet/public/set and input coordinates of the stars can be found in PublicDataSet/public/input.
To test the example run the Mapping Stars.py file and input the number of example you want to test.