CHI '99 Extended Abstracts on Human Factors in Computing Systems - CHI '99 1999
DOI: 10.1145/632716.632726
|View full text |Cite
|
Sign up to set email alerts
|

CoMedi

Abstract: CoMedi is a mediaspace prototype that uses computer vision to provide new solutions to the problems of visual discontinuity and privacy. CoMedi includes a robust face tracker based on cooperation of multiple vision techniques, a tele-exploration tool based on a multi-resolution fovea, and new privacy filter using eigen-space coding.

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1

Citation Types

0
2
0

Year Published

2001
2001
2011
2011

Publication Types

Select...
3
2
1

Relationship

0
6

Authors

Journals

citations
Cited by 10 publications
(2 citation statements)
references
References 7 publications
0
2
0
Order By: Relevance
“…However, it would be ideal to try and find a "universal" model that could classify the level of engagement of a user who was not part of the training set. We believe that an excellent step in that direction would be the work of Coutaz et al [4]. We could create a basis set of images of facial components from many different users using PCA (principal components analysis), a technique that aims to represent as much of the original variance of a set of data as compactly as possible , then encode user faces using this basis, and use the encodings to create an SVM model.…”
Section: Discussionmentioning
confidence: 99%
See 1 more Smart Citation
“…However, it would be ideal to try and find a "universal" model that could classify the level of engagement of a user who was not part of the training set. We believe that an excellent step in that direction would be the work of Coutaz et al [4]. We could create a basis set of images of facial components from many different users using PCA (principal components analysis), a technique that aims to represent as much of the original variance of a set of data as compactly as possible , then encode user faces using this basis, and use the encodings to create an SVM model.…”
Section: Discussionmentioning
confidence: 99%
“…An early paper by Coutaz et al [4] suggested integration with a more mathematical fieldcomputer vision. Specifically, it presented techniques for providing awareness of the video of a user to remote collaborators without violating privacy concerns.…”
Section: Introductionmentioning
confidence: 99%