Title :
Face Segmentation under Unconstrained Scenes
Author_Institution :
Inst. of Photonics & Commun., Nat. Kaohsiung Univ. of Appl. Sci.
Abstract :
In this paper, an efficient approach by combining the novel wavelet-based feature template, the support vector machine (SVM) classifier, and the wavelet entropy filtering is presented to robustly detect and segment human face image under complex background. Moreover, a face detection measure (FDM) criterion based on the distance between the expected and the detected eye-mouth triangle circumscribed circle areas is introduced to validate the performance of precise face segmentation
Keywords :
face recognition; feature extraction; filtering theory; image classification; image segmentation; object detection; support vector machines; wavelet transforms; FDM; SVM classifier; eye-mouth triangle; face detection measure; face segmentation; support vector machine; unconstrained scene; wavelet entropy filtering; wavelet-based feature template; Area measurement; Entropy; Face detection; Filtering; Humans; Image segmentation; Layout; Robustness; Support vector machine classification; Support vector machines;
Conference_Titel :
Multimedia and Expo, 2006 IEEE International Conference on
Conference_Location :
Toronto, Ont.
Print_ISBN :
1-4244-0366-7
Electronic_ISBN :
1-4244-0367-7
DOI :
10.1109/ICME.2006.262910
