Title :
Sensitivity evaluation of SNoW and SVM based face detection methods
Author :
Liu, Jialin ; Rautiainen, Mika ; Ojala, Timo
Author_Institution :
MediaTeam Oulu Group, Oulo Univ.
Abstract :
This paper studies the performance sensitivity of the variations in different factors on SNoW and SVM based face detection techniques. These factors include combinations of training set, pixel transitions in sliding window, pre-processing methods (rotation, scaling, light correction, masking filter), and bootstrapping in negative training patterns. It was found that performance of the two detection methods is significantly sensitive to these factors
Keywords :
face recognition; sensitivity; support vector machines; SNoW based face detection; SVM based face detection; negative training patterns; performance sensitivity; pixel transitions; preprocessing methods; sensitivity evaluation; sliding window; training set; Face detection; Filters; Histograms; Image databases; Lighting; Performance evaluation; Pixel; Snow; Support vector machines; Testing;
Conference_Titel :
Multi-Media Modelling Conference Proceedings, 2006 12th International
Conference_Location :
Beijing
Print_ISBN :
1-4244-0028-7
DOI :
10.1109/MMMC.2006.1651367
