Research and Publications
Research Project :- Automatic Human Face Detection and Localization
Automatic Detection and Localization of Human Faces in scenes with no prior information about scale, orientation and viewpoint.
Here are some image sequences in MPEG showing the results of the face detector. No visual tracking were used.
- Sequence 1 Image sequence of a face moving in strong illumination.
- Sequence 2 Image sequence of a face moving in saturated camera constrast.
- Sequence 3 Simultaneous detection of more than one face in the image.
- Sequence 4 Detection of a hand drawn face.
Publications on FTP
Compressed Postscript (.ps.Z) files available on the SVR ftp site.
- Enhancing Human Face Detection using Motion and Active Contours. In Proceedings 3rd Asian Conference on Computer Vision, vol. 1, pp 515-522, Hong Kong, 1998.
- A Probabilistic Framework for Perceptual Grouping of Features for Human Face Detection. In Proceedings 2nd International Conference on Automatic Face and Gesture Recognition, pp 16-21, Vermont, 1996.
- Detection of Human Faces under Scale, Orientation and Viewpoint Variations. In Proceedings 2nd International Conference on Automatic Face and Gesture Recognition, pp 295-300, Vermont, 1996.
- Scale and Orientation Invariance in Human Face Detection. In Proceedings 7th British Machine Vision Conference, vol. 2, pp 745-754, Edinburgh, 1996.
- Finding Initial Estimates of Human Face Location. In Proceedings 2nd Asian Conference on Computer Vision, vol. 3, pp 514-518, Singapore, 1995.
- Towards an Automatic Human Face Localization System. In Proceedings 6th British Machine Vision Conference, vol. 2, pp 701-710, Birmingham, 1995.