Skip to content

Latest commit

 

History

History
61 lines (32 loc) · 2.11 KB

README (1).md

File metadata and controls

61 lines (32 loc) · 2.11 KB

EAD2019 challenge

Video endoscopy image data and labels

[1] First training set data

    Release date: 15th November 2018
  • Number of video frames/images = 887
    • Data institutions: JRC, UK/ICL, France/ APHP, France/ UHV, Switzerland/Botkin Clinical City Hospital, Russia
    • Data modalities: White light/Fluorescence/Narrow-band imaging
    • Data resolutions: Standard HD/Full HD/high magnification and others
    • Tissue types: Oesophagus, stomach, pylorus, rectum, anus , colon, bladder...

[2] Second training set data

TO BE DECLARED!!!

[3] Test data release

TO BE DECLARED (Tentative in Febraury)

See for updated at EAD2019

Annotation

  • Annotation provided are for 7 classes and consists of intotal 9352 bounding boxes (bbox)

  • Distribution of class specific bboxes in the provided dataset is presented in the bar graph

    image

  • Bounding box labels has been provided in the below format:

      <object-class> <x> <y> <width> <height> : Ideally, <0> <0> <1> <1>

  <object-class> integer values from 0 to 6
  <x> = <abs x>/<image_width>, <y> = <abs y>/<image_height> they correspond to annotators box's centres
  <width>: annotator box normalised width 
  <height> annotator box normalised height

  • Annotation converter to corresponding non-normalized format as csv file is provided as software tools

  • class labels --> 0: specularity ..... 6:instrument

Software tools

Terms and conditions of data usage:

[1] All data provided in this challenge is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License

[2] Image data can be used for research purposes only. Utilization of any data for a product or (self) publication without consent and proper legal agreement from the EAD2019 organisers is forbidden!!!

[3] See EAD2019 website for challenge participation rules