Dear organizers,
I was hoping you could shed some light on the csv files associated with the skills task.
For each full video, there is a "calot" and "dissection" video with its associated skill annotation file (named "Calot_Skill.csv" and "Dissection_Skill.csv").
So I assume for each of the "calot" / "dissection" video snippets we are supposed to predict the skill presented in the associated csv.
However, there is also a third csv file, called "Skill.csv" for each full video. Are we supposed to also predict the skill presented in this file? What does the file refer to?
First, I thought the skill annotations in this file are a rounded average of the "calot" and "dissection" skill but that doesn't seem to be the case looking at video number 18.
An explanation would be highly appreciated.
Thank you!
Created by isatwi isatwi Hello,
as mentioned in the data description/ReadMe, the Calot_Skill.csv and Dissection_Skill.csv refer to the extracted videos. The Skill.csv prefers to a skill assessment performed on the original video. We will clarify this in the descriptions.
Regards,
Sebastian