UMD Sushi-Making Dataset

There are total 4 actors: a1, a2, a3, a4 with 12 actions [cleaning], [cutting], [drinking], [flipping], [peeling], [picking], [pouring], [pressing], [sprinkling], [stirring], [tossing], [turning].

Download the dataset [tar.gz] for each actor:

Synced MOCAP data are available upon request.

Questions? Please contact cteo "at" umd dot edu