Presentation skills estimation based on video and kinect data analysis

Vanessa Echeverría, Allan Avendaño, Katherine Chiluiza, Aníbal Vásquez, Xavier Ochoa

Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

40 Citations (Scopus)

Abstract

This paper identifies, by means of video and Kinect data, a set of predictors that estimate the presentation skills of 448 individual students. Two evaluation criteria were predicted: eye contact and posture and body language. Machine-learning evaluations resulted in models that predicted the perfor- mance level (good or poor) of the presenters with 68% and 63% of correctly classified instances, for eye contact and postures and body language criteria, respectively. Furthermore, the results suggest that certain features, such as arms movement and smoothness, provide high significance on predicting the level of development for presentation skills. The paper finishes with conclusions and related ideas for future work.

Original languageEnglish
Title of host publicationMLA 2014 - Proceedings of the 2014 ACM Multimodal Learning Analytics Workshop and Grand Challenge, Co-located with ICMI 2014
PublisherAssociation for Computing Machinery (ACM)
Pages53-60
Number of pages8
ISBN (Electronic)9781450304887
DOIs
Publication statusPublished - 12 Nov 2014
Externally publishedYes
EventMultimodal Learning Analytics Workshop and Grand Challenges 2014 - Istanbul, Türkiye
Duration: 12 Nov 201412 Nov 2014
Conference number: 3rd
http://icmi.acm.org.ezproxy.lib.monash.edu.au/2014/

Workshop

WorkshopMultimodal Learning Analytics Workshop and Grand Challenges 2014
Abbreviated titleMLA 2014
Country/TerritoryTürkiye
CityIstanbul
Period12/11/1412/11/14
OtherEvent = 16th ACM International Conference on Multimodal Interaction, ICMI 2014 - Istanbul, Turkey
Duration: Nov 12 2014 → Nov 16 2014

Proceedings title = ICMI 2014 - Proceedings of the 2014 International Conference on Multimodal Interaction
Internet address

Keywords

  • Multimodal
  • Presentation skills
  • Video features

Cite this