This project focuses on image segmentation using a custom dataset and state-of-the-art convolutional neural networks (Unet). We apply data augmentation techniques and leverage the Albumentations library to enhance model performance.

Custom Dataset Class: We developed a custom dataset class tailored for image-mask pairs, ensuring efficient data organization.

Data Augmentation: Utilizing Albumentations, we applied segmentation augmentation to both images and masks, improving model robustness.

Visual Insights: We visually explored the dataset by plotting image-mask pairs, gaining valuable insights.

State-of-the-Art Model: Loaded a pretrained Unet model using the Segmentation Models PyTorch library for top-tier performance.

Training Functionality: Created train and evaluator functions to efficiently train the model and evaluate its performance.

Follow these instructions to get a copy of the project up and running on your local machine for development and testing purposes.

Python 3.x

PyTorch

Clone the repository:

git clone https://github.com/TLILIFIRAS/Image-Segmentation-With-PyTorch.git cd Image-Segmentation-With-PyTorch