goto contents

Kristina Yordanova     Frank Krüger     Thomas Kirste

Semantic Annotation for the CMU-MMAC Dataset (Version 2) : [research data]

University of Rostock, 2018

Abstract: Objective: To create semantic annotation of the Carnegie Mellon University Multi-Modal Activity Database (CMU-MMAC) grand challenge kitchen dataset, which is often cited but, due to missing and incomplete annotation, almost never used. Changes to Earlier Version: (1) Changes to the starting times and the actual plan sequences. The previous version contains errors in the plan sequences. (2) The dataset now also contains the models that were used to validate the annotation plan sequences. The models were also used for semantic reasoning on the environment properties.


data publication   free access