This repository contains the code for semantic segmentation on the Lung Segmentation dataset using TensorFlow 2.0 framework. The following models are used:
Models to be used in future:
- RESUNET
- DEEPLABV3+
- more...
Download the MontgomerySet dataset.
The Montgomery dataset is used for this for training the UNET architecture. The dataset contains the 138 pairs of images and masks. All of these images are of different shapes and contains a variety of skin lesions.
| Original Image | Left Lung Mask | Right Lung Mask |
|---|---|---|
 |
 |
 |