Segmentation-ChestXDet

Semantic segmentation of diseases using the ChestX-Det dataset. This project includes the complete pipeline for creating the data loader, training, and testing the model. The models used are UNet and UNet++. The segmentation-models-pytorch (https://github.com/qubvel/segmentation_models.pytorch) library is used to produce the results.