I am following this tutorial on image segmentation on the TensorFlow website.
The website uses its own labeled images for the tutorial, so the images have data that says which pixels are a part of the object, which ones border the object, and which pixels aren't part of the object.
This tutorial uses the Oxford-IIIT Pet Dataset, created by Parkhi et al. The dataset consists of images of 37 pet breeds, with 200 images per breed (~100 each in the train and test split). Each image includes the corresponding labels, and pixel-wise masks. The masks are class-labels for each pixel. Each pixel is given one of three categories :
- Class 1: Pixel belonging to the pet.
- Class 2: Pixel bordering the pet.
- Class 3: None of the above/ Surrounding pixel.
In my case, I have unlabelled images, so I cannot currently perform image segmentation with my images. Which approach should I use to label my images for image segmentation?