A simple multiclass segmentation tutorial on the Oxford-IIIT Pet dataset using the U-Net architecture.
Video explaination: https://youtu.be/afqf_sxDyiY
Download the dataset: https://www.robots.ox.ac.uk/~vgg/data/pets
The images given below are in the sequence: (1) Input Image, (2) Ground Truth, (3) Predicted Mask
For more follow me on: