What is MOT-17?
MOT-17 is a multi-object tracking dataset benchmark that is widely cited and used for research papers to demonstrate and compare performance. It is a composed of different video segments with densely populated pedestrian scenes and includes annotations generated from three different detectors with the corresponding truth annotations.
View an example dataset here.
Comments
One response to “MOT-17 Explained Simply”
Hi, this is a comment.
To get started with moderating, editing, and deleting comments, please visit the Comments screen in the dashboard.
Commenter avatars come from Gravatar.