CSR-STNet

Implementation of paper Colo-segment Recognition in Colonoscopy Videos via SpatioTemporal Network

Requirements

• torch==2.0.1
• torchvision==0.15.2

Instructions

Alter configs file to change model and training configurations.
Use command CUDA_VISIBLE_DEVICES=0 python main.py --config_path ./configs/default.yml to train CSR-STNet.

Acknowledgements

• https://github.com/LTContext/LTContext
• https://github.com/xjgaocs/Trans-SVNet