Skip to content

Sample video and possibly data for WearSys 2020 paper Challenges in Measuring Partner Dancing Skills via Wearable Accelerometers.

Notifications You must be signed in to change notification settings

elleros/dance-accelerometry

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 

Repository files navigation

Social partner dancing is a fun but challenging activity requiring different motion related skills. Common criteria used by professionals to assess the quality of this type of dancing fall in the categories of timing, technique and teamwork (often referred to as the "3 Ts") and variety of motion (i.e. "moves"). We focus on the teamwork and variety skills for practitioners of a type of Swing dancing called Balboa. Our dataset consists of the wearable accelerometer data collected from the participants to 3 different Balboa social dance contests. The contests were judged by panels of professional dancers. Later, some of those professional dancers evaluated the skills of each participant by watching video recordings of the contests. We propose four novel measures for teamwork and motion variety and we evaluate them versus the expert assessments and also activity based labels. Our preliminary results show that the measures can be useful for activity recognition, but not for skill assessment.

Link to full paper

Demo clip of one of the contests used to collect the data

Demo Clip

About

Sample video and possibly data for WearSys 2020 paper Challenges in Measuring Partner Dancing Skills via Wearable Accelerometers.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published