Tutorial: OpenPose for Linguists

3 October 2019 14:00 — 17:30
Hamburg University, Germany

Course Description

When working with multimodal data, most annotation steps are done manually. OpenPose from Carnegie Mellon University opens the door to analyzing the visual domain without becoming an expert in Computer Vision: In essence, OpenPose determines the position of joints and returns them as time-series data. This tutorial shows you how to work with OpenPose in order to detect handedness, headshake & more and how to feed the results into your annotation, independent of whether you work with sign or gesture.

Marc Schulder
Research Associate in Computational Linguistics

My research interests include natural language processing, sign languages and sentiment analysis.