Zooming into Abnormal Events in Video Conferencing

Shmuel Horowitz, Dima Kagan, Galit Fuhrmann Alpert, Michael Fire

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


Video conferencing (VC) has become increasingly popular, bringing new challenges in privacy and security, one notable example is of Zoombombing. Furthermore, other issues related to VC usage have emerged, such as keeping students involved. Identifying abnormal segments in VC meetings in vast data is a challenging task. Here, we introduce a novel algorithm to detect such anomalies in VC automatically. By analyzing publicly available VC recordings, our algorithm tracks and analyzes changes in participants' facial expressions to identify and quantity overall meeting climate changes. We demonstrate performance of 92.3% precision in anomaly detection on the collected dataset. Our model offers a pioneering solution for recognizing abnormal events in VC meetings.

Original languageEnglish
Title of host publicationProceedings - 2023 IEEE International Conference on Big Data, BigData 2023
EditorsJingrui He, Themis Palpanas, Xiaohua Hu, Alfredo Cuzzocrea, Dejing Dou, Dominik Slezak, Wei Wang, Aleksandra Gruca, Jerry Chun-Wei Lin, Rakesh Agrawal
PublisherInstitute of Electrical and Electronics Engineers
Number of pages6
ISBN (Electronic)9798350324457
StatePublished - 1 Jan 2023
Event2023 IEEE International Conference on Big Data, BigData 2023 - Sorrento, Italy
Duration: 15 Dec 202318 Dec 2023

Publication series

NameProceedings - 2023 IEEE International Conference on Big Data, BigData 2023


Conference2023 IEEE International Conference on Big Data, BigData 2023


  • Anomaly detection in video
  • Video conference

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Networks and Communications
  • Computer Science Applications
  • Information Systems
  • Information Systems and Management
  • Safety, Risk, Reliability and Quality


Dive into the research topics of 'Zooming into Abnormal Events in Video Conferencing'. Together they form a unique fingerprint.

Cite this