Info: Zenodo’s user support line is staffed on regular business days between Dec 23 and Jan 5. Response times may be slightly longer than normal.

Published March 13, 2023 | Version 0.0.1
Dataset Open

StereoMIS

Creators

  • 1. Universtität Bern
  • 2. Intuitive Surgical

Description

StereoMIS

StereoMIS is a dataset for Simultaneous Localization and Mapping (SLAM) in endoscopic surgery. The dataset has been recorded using the da Vinci Xi surgical robot, capturing a stereo video stream and camera forward kinematics. It consists of three in-vivo porcine subjects with 11 surgical sequences containing breathing, tool motion, and tissue deformations. Each folder contains:

  • a video file with vertically stacked video feeds from the stereo endoscopes
  • automatically generated tool masks for the left video frames
  • camera forward kinematics synchronized with the video feed
  • csv-file containing start and stop frames of the relevant sequence

 

  • If you use the dataset in your research, please cite our paper:
@article{hayoz2023pose,
title={Learning how to robustly estimate camera pose in endoscopic videos},
author={Michel Hayoz, Christopher Hahne, Mathias Gallardo, Daniel Candinas, Thomas Kurmann, Max Allan, Raphael Sznitman},
journal={International Journal of Computer Assisted Radiology and Surgery 2023}
doi={https://doi.org/10.1007/s11548-023-02919-w}
}

Release notes:

  • added missing P3
  • added missing start and stop frames
  duration total trajectory length frame-rate frame-rate extracted split
P1 5min 19s 1323 mm 60 fps 30 fps train
P2_0 3min 28s 434 mm 52.6 fps 26.3 fps test
P2_1 2min 36s 1155 mm 54.4 fps 27.2 fps test
P2_2 2min 37s 736 mm 54.4 fps 27.2 fps test
P2_3 2min 35s 302 mm 54.4 fps 27.2 fps test
P2_4 2min 37s 1824 mm 54.4 fps 27.2 fps test
P2_5 3min 18s 1671 mm 54.2 fps 27.1 fps test
P2_6 52s 205 mm 26.1 fps 26.1 fps test
P2_7 2min 26s 725 mm 37.0 fps 37.0 fps test
P2_8 3min 55s 950 mm 39.0 fps 39.0 fps test
P3 2min 58s 545 mm 60.0 fps 30.0 fps test


This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

Files

StereoMIS_0_0_1.zip

Files (11.2 GB)

Name Size Download all
md5:27d5389e17100ec415445cb975df791b
282 Bytes Download
md5:e014067d348ad81fff94ac039b97abe8
11.2 GB Preview Download