Summary
2011 International Symposium on Nonlinear Theory and Its Applications
2011
Session Number:B4L-B
Session:
Number:B4L-B1
Multi-resolution-type AAM for Low-resolution Images
Satoru Suzuki, Yasue Mitsukura,
pp.529-532
Publication Date:2011/9/4
Online ISSN:2188-5079
DOI:10.34385/proc.45.B4L-B1
PDF download (349.1KB)
Summary:
The purpose of this paper is to extract small faces from images. When a mobile robot tries to find a person from a far distance, the image resolution will be degraded largely, which makes face tracking difficult. Therefore, we present the multi-resolution-type active appearance model (AAM) robust to degraded facial images. In the proposed method, both high-resolution and downsampled low-resolution images are used together to train an AAM. The shape matching performance of multi-resolution-type AAM is evaluated by using images from low-resolution to high-resolution. The simulation results show that multiresolution-type AAM has the possibility for coping with variety of image resolutions.