This challenge is divided into two categories. Details on each category are provided below

Category 1: Surgical tool classification¶

This category will require the teams to train multi-label fully supervised models. The model should classify all tools present within each frame of the video clips in the test set by training on the tool presence labels provided in the training set.

Category 2: Surgical tool classification and localization¶

This category will require the teams to train weakly supervised models. The model should localize (with bounding boxes) and classify the tools present within each frame of the video clips in the test set by training on the tool presence labels provided in the training set.