Vibashan / AP-MTL

Official Pytorch Codebase for AP-MTL [ICRA 2020]

AP-MTL

This repository contains the official implementation of the paper "AP-MTL: Attention Pruned Multi-task Learning Model for Real-time Instrument Detection and Segmentation in Robot-assisted Surgery" preprint and in ICRA 2020 Proceedings

The baseline implementation of SSD is adopted from
SSD: https://github.com/qijiezhao/pytorch-ssd

Dataset

2017 Robotic Instrument Segmentation Challenge and paper

Citation

If you use this code for your research, please cite our paper.

@inproceedings{islam2020ap,
  title={Ap-mtl: Attention pruned multi-task learning model for real-time instrument detection and segmentation in robot-assisted surgery},
  author={Islam, Mobarakol and Vibashan, VS and Ren, Hongliang},
  booktitle={2020 IEEE International Conference on Robotics and Automation (ICRA)},
  pages={8433--8439},
  year={2020},
  organization={IEEE}
}