Hierarchical wavelet networks for facial feature localization
(2002) 5th IEEE International Conference on Automatic Face Gesture Recognition, FGR 2002 p.125-130- Abstract
We present a technique for facial feature localization using a two-level hierarchical wavelet network. The first level wavelet network is used for face matching, and yields an affine transformation used for a rough approximation of feature locations. Second level wavelet networks for each feature are then used to fine-tune the feature locations. Construction of a training database containing hierarchical wavelet networks of many faces allows features to be detected in most faces. Experiments show that facial feature localization benefits significantly from the hierarchical approach. Results compare favorably with existing techniques for feature localization.
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/f612c91a-c5a3-4810-a013-51d0dc1cc171
- author
- Feris, Rogério Schmidt ; Gemmell, Jim ; Toyama, Kentaro and Krüger, Volker LU
- publishing date
- 2002-01-01
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- host publication
- Proceedings - 5th IEEE International Conference on Automatic Face Gesture Recognition, FGR 2002
- article number
- 1004143
- pages
- 6 pages
- publisher
- IEEE - Institute of Electrical and Electronics Engineers Inc.
- conference name
- 5th IEEE International Conference on Automatic Face Gesture Recognition, FGR 2002
- conference location
- Washington, DC, United States
- conference dates
- 2002-05-20 - 2002-05-21
- external identifiers
-
- scopus:84905392651
- ISBN
- 0769516025
- 9780769516028
- DOI
- 10.1109/AFGR.2002.1004143
- language
- English
- LU publication?
- no
- id
- f612c91a-c5a3-4810-a013-51d0dc1cc171
- date added to LUP
- 2019-07-08 21:24:08
- date last changed
- 2022-04-10 20:04:56
@inproceedings{f612c91a-c5a3-4810-a013-51d0dc1cc171, abstract = {{<p>We present a technique for facial feature localization using a two-level hierarchical wavelet network. The first level wavelet network is used for face matching, and yields an affine transformation used for a rough approximation of feature locations. Second level wavelet networks for each feature are then used to fine-tune the feature locations. Construction of a training database containing hierarchical wavelet networks of many faces allows features to be detected in most faces. Experiments show that facial feature localization benefits significantly from the hierarchical approach. Results compare favorably with existing techniques for feature localization.</p>}}, author = {{Feris, Rogério Schmidt and Gemmell, Jim and Toyama, Kentaro and Krüger, Volker}}, booktitle = {{Proceedings - 5th IEEE International Conference on Automatic Face Gesture Recognition, FGR 2002}}, isbn = {{0769516025}}, language = {{eng}}, month = {{01}}, pages = {{125--130}}, publisher = {{IEEE - Institute of Electrical and Electronics Engineers Inc.}}, title = {{Hierarchical wavelet networks for facial feature localization}}, url = {{http://dx.doi.org/10.1109/AFGR.2002.1004143}}, doi = {{10.1109/AFGR.2002.1004143}}, year = {{2002}}, }