Image Segmentation Using U-Net Model

• Understand the Segmentation Dataset and write a custom dataset class for the image-mask dataset.

• Apply segmentation augmentation to augment images as well as their masks. For image-mask augmentation, albumentation library is used.

• Plot the image-mask pair.

• Load a pre-trained state-of-the-art CNN model for the segmentation problem i.e., U-Net using the segmentation_models_pytorch library.

• Create a train and evaluator function which will be helpful to write a training loop.