Finding Time Together: Detection and Classification of Focused Interaction in Egocentric Video

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)
164 Downloads (Pure)

Abstract

Focused interaction occurs when co-present individuals, having mutual focus of attention, interact by establishing face-to-face engagement and direct conversation. Face-toface engagement is often not maintained throughout the entirety of a focused interaction. In this paper, we present an online method for automatic classification of unconstrained egocentric (first-person perspective) videos into segments having no focused interaction, focused interaction when the camera wearer is stationary and focused interaction when the camera wearer is moving. We extract features from both audio and video data streams and perform temporal segmentation by using support vector machines with linear and non-linear kernels. We provide empirical evidence that fusion of visual face track scores, camera motion profile and audio voice activity scores is an effective combination for focused interaction classification.
Original languageEnglish
Title of host publication2017 IEEE International Conference on Computer Vision Workshop (ICCVW)
PublisherIEEE
Pages2322-2330
Number of pages9
ISBN (Electronic)9781538610343
ISBN (Print)9781538610350
DOIs
Publication statusPublished - 23 Jan 2018
EventIEEE International Conference on Computer Vision Workshops - Venice Convention Centre, Venice, Italy
Duration: 22 Oct 201729 Oct 2017
http://iccv2017.thecvf.com/

Publication series

NameProceedings - 2017 IEEE International Conference on Computer Vision Workshops, ICCVW 2017
Volume2018-January

Conference

ConferenceIEEE International Conference on Computer Vision Workshops
Abbreviated titleICCV 2017
CountryItaly
CityVenice
Period22/10/1729/10/17
Internet address

Keywords

  • Cameras
  • Face
  • Feature extraction
  • Tracking
  • Visualization
  • Legged locomotion

Fingerprint Dive into the research topics of 'Finding Time Together: Detection and Classification of Focused Interaction in Egocentric Video'. Together they form a unique fingerprint.

Cite this