Skip to Main Content
ASME Press Select Proceedings

International Conference on Instrumentation, Measurement, Circuits and Systems (ICIMCS 2011)

Chen Ming
Chen Ming
Search for other works by this author on:
No. of Pages:
ASME Press
Publication date:

The pose variation and one sample in facial image degrades face recognition technologies considerably, to solve the problem, this paper present a one sample face recognition across pose method.We use FHMM (Fourier-HMM) to estimate the pose of the input non-frontal face,then we can obtain pose-HMM. Linear regression is used to learn the approximate linear mapping between non-frontal face and its frontal face image. Then the input non-frontal face can be transformed to its virtual frontal face following pose estimation resultit's called pose correction. For the case of one sample, this paper obtains additional virtual samples by projecting the single sample into different mappings. Finally, HMM is used as a classifier to recognize the faces across pose. Experimental results in Weizmann database show that the method can achieve good recognition accuracy.

This content is only available via PDF.
Close Modal
This Feature Is Available To Subscribers Only

Sign In or Create an Account

Close Modal
Close Modal