Smile Detection by Boosting Pixel Differences Caifeng Shan, Member, IEEE IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 21, NO. 1, JANUARY 2012
Transcript
Slide 1
Caifeng Shan, Member, IEEE IEEE TRANSACTIONS ON IMAGE
PROCESSING, VOL. 21, NO. 1, JANUARY 2012
Slide 2
INTRODUCTION METHOD EXPERIMENTS
Slide 3
INTRODUCTION METHOD EXPERIMENTS
Slide 4
Most of the existing works have been focused on analyzing a set
of prototypic emotional facial expressions Using the data collected
by asking subjects to pose deliberately these expressions In this
paper, we focus on smile detection in face images captured in
real-world scenarios
Slide 5
Slide 6
INTRODUCTION METHOD EXPERIMENTS
Slide 7
BOOSTING PIXEL DIFFERENCES S. Baluja and H. A. Rowley, Boosting
set identification performance,Int. J. Comput. Vis., vol. 71, no.
1, pp. 111119, 2007 Baluja introduced to use the relationship
between two pixels intensities as features.
Slide 8
they used five types of pixel comparison operators (and their
inverses):
Slide 9
The binary result of each comparison, which is represented
numerically as 1 or 0, is used as the feature. Thus, for an image
of pixels, there are or 3312000 pixel- comparison features
Slide 10
Instead of utilizing the above comparison operators, we propose
to use the intensity difference between two pixels as a simple
feature For an image of 24*24 pixels, there are or 331200 features
extracted
Slide 11
AdaBoost ( Adaptive Boosting ) AdaBoost learns a small number
of weak classifiers whose performance is just better than random
guessing and boosts them iteratively into a strong classifier of
higher accuracy the weak classifier consists of feature (i.e., the
intensity difference),threshold, and parity indicating the
direction of the inequality sign as follows:
Slide 12
Slide 13
Slide 14
INTRODUCTION METHOD EXPERIMENTS
Slide 15
Data Database : GENKI4K consists of 4000 images (2162 smile and
1828 nonsmile) In our experiments, the images were converted to
grayscale the faces were normalized to reach a canonical face of
48*48 pixels