CCG Home | People | Research | Publications | Paper of the Week | Resources | Contact

Learning Visual Models Of Social Engagement

Learning Visual Models of Social Engagement

Brad Singletary and Thad Starner
Contextual Computing Group


The wearable face recognition and detection problem differs significantly from the classical problem and its solutions. This is due to large variations in lighting, scene, and camera positions that occur in everyday situations. The problem is currently compounded by computational limitations induced by useability and mobility requirements of such a device. Prior work has only proved the feasibility of constructing such a device.  We propose a formal study of the space of data acquireable from such a device and the interfaces for its acquisition.


A paper on using a wearable to understand when I'm in a conversation
A paper on building interfaces for wearable face recognition


Current embodiment(coming very soon)

An earlier capture vest that enables mobile audio/video capture

Short example movie (windows/linux DV encoded AVI - 60MB) that a user captured using the vest.
A Montage of several extracted faces  
A peek at the mean face and first 20 color Eigenfaces computed from that set of faces after cropping them by the social constraint method and reducing them to 31x22x3.


A class project I did in 2000 on using the frequency content of the face to segment it within social interaction distances.
Prior wearable face detection work...
MIT Augmented Reality Project
Augmented Reality Through Wearable Computing
LandWarrior Project

Contents Copyright 2000 Brad Singletary and Georgia Tech
Contextual Computing Group

CCG Home | People | Research | Publications | Paper of the Week | Resources | Contact

[GVU Center] [College Of Computing] [Georgia Tech]