[OpenPose][1] [arXiv:1812.08008, arXiv:1704.07809] detects positions of the body, hand, face and foot of multiple people. OpenPose can be applied to videos to track people’s movements. OpenPose detects 135 key points of the human body in total. OpenPose is mindblowing in what it can do, but it might be a daunting task for many social scientists to get it to work. To make it easier to use OpenPose, we created two things: 1. An *OpenPose for dummies* manual (RepgenDeJongeHoekstra_Manual[...].pdf) 2. An R-script that transforms the OpenPose output (JSON files) for the body key points** to a correctly ordered csv file for max. 2 persons*** (openpose2R[...].R) Furthermore, we added the following things to the OSF folder: 3. A poster presentation about this package (DeJongeHoekstraRepgen_Poster[...].pdf) 4. A demo video that you can use to try out OpenPose for the first time (Openpose_demo.mp4) ** A script that is able to correctly order the output of the hand and face key points will become available in the future. *** A script that is able to correctly order the output of more than 2 persons will become available in the future. You can use the script on the OpenPose output of a video with more than 2 persons, but only the output of the two persons that are listed first in the first JSON file will appear in the csv file. [1]: https://github.com/CMU-Perceptual-Computing-Lab/openpose