PAUL VIOLAMicrosoft Research - PowerPoint PPT Presentation

1 / 20
About This Presentation
Title:

PAUL VIOLAMicrosoft Research

Description:

Given example images (x1, y1), . . . , (xn , yn ) where yi = 0, 1 for negative ... are tilted up to about 15 degrees in plane and about 45 degrees out of plane. ... – PowerPoint PPT presentation

Number of Views:37
Avg rating:3.0/5.0
Slides: 21
Provided by: Kai71
Category:

less

Transcript and Presenter's Notes

Title: PAUL VIOLAMicrosoft Research


1
Robust Real-Time Face Detection
  • PAUL VIOLA Microsoft Research
  • MICHAEL J.JONES Mitsubishi Electric Research
    Lab.

2
OUTLINE
  • INTRODUCTION
  • RELATED WORK
  • FEATURES
  • Adaptive Boosting
  • Cascade of Classifier
  • RESULTS
  • CONCLUSIONS

3
INTRODUCTION
  • INTRODUCTION
  • FEATURES
  • Adaptive Boosting
  • Cascade of Classifier
  • RESULTS
  • This face detection system is most clearly
    distinguished from previous approaches in its
    ability to detect faces extremely fast.
  • Operating on 384 by 288 pixel images, faces are
    detected at 15 frames per second on a
    conventional 700 MHZ Intel Pentium III.
  • Achieves higher frame rates by working only with
    the information present in a single gray scale
    image.

4
INTRODUCTION
  • INTRODUCTION
  • FEATURES
  • Adaptive Boosting
  • Cascade of Classifier
  • RESULTS
  • 3 main contributions of this system.
  • A new image representation called integral
    image that allows very fast feature evaluation.
  • Selecting a small number of important features
    from a huge library of potential features using
    AdaBoost (Adaptive Boosting).
  • Combining classifiers in a cascade structure
    which increases the speed of the detector by
    focusing on promising regions of the image.

5
FEATURES
  • INTRODUCTION
  • FEATURES
  • Adaptive Boosting
  • Cascade of Classifier
  • RESULTS
  • Integral image
  • The value of the integral image at point (x,y) is
    the
  • sum of all the pixels above and to the left.

6
FEATURES
  • INTRODUCTION
  • FEATURES
  • Adaptive Boosting
  • Cascade of Classifier
  • RESULTS
  • Rectangle features
  • The sum of the pixels which lie within the white
  • rectangle are subtracted from the sum of pixels
    in
  • the grey rectangles.

7
FEATURES
  • INTRODUCTION
  • FEATURES
  • Adaptive Boosting
  • Cascade of Classifier
  • RESULTS
  • Rectangle features

8
FEATURES
  • INTRODUCTION
  • FEATURES
  • Adaptive Boosting
  • Cascade of Classifier
  • RESULTS
  • Scanning the image
  • The detector is scanned across the input image at
    multiple scales and locations.
  • Our detector scans the input at many scales
    starting at the base size of 24 24 pixels, a
    384 by 288 pixel image is scanned at 12 scales
    each a factor of 1.25 larger than the last.
  • The detector is also scanned across location.
    Subsequent locations are obtained by shifting the
    window some number of pixels ?.

9
Adaptive Boosting
  • INTRODUCTION
  • FEATURES
  • Adaptive Boosting
  • Cascade of Classifier
  • RESULTS
  • Given a feature set and a training set of
    positive and negative images, adaptive boosting
    could be used to learn a classification function.
  • Each feature is used as a weak classifier.
  • A weak classifier ( h (x, f, p, ?)) thus consists
    of a feature ( f ), a threshold (?) and a
    polarity (p) indicating the direction of the
    inequality

10
Adaptive Boosting
  • INTRODUCTION
  • FEATURES
  • Adaptive Boosting
  • Cascade of Classifier
  • RESULTS
  • Algorithm
  • Given example images (x1, y1), . . . , (xn , yn )
    where yi 0, 1 for negative and positive
    examples respectively.
  • Initialize weights for yi 0, 1
    respectively, where m and l are the number of
    negatives and positives respectively.

11
Adaptive Boosting
  • INTRODUCTION
  • FEATURES
  • Adaptive Boosting
  • Cascade of Classifier
  • RESULTS
  • Algorithm

12
Adaptive Boosting
  • INTRODUCTION
  • FEATURES
  • Adaptive Boosting
  • Cascade of Classifier
  • RESULTS
  • The first two features selected by AdaBoost.
  • The first feature measures the difference in
    intensity between the region of the eyes and a
    region across the upper cheeks.
  • The second feature compares the intensities in
    the eye regions to the intensity across the
    bridge of the nose.

13
Cascade of Classifier
  • INTRODUCTION
  • FEATURES
  • Adaptive Boosting
  • Cascade of Classifier
  • RESULTS
  • Within an image, most sub-images are non-face
    instances.
  • Use smaller and efficient classifiers to reject
    many negative examples at early stage while
    detecting almost all the positive instances.
  • Simpler classifiers are used to reject the
    majority of sub-windows.
  • More complex classifiers are used at later stage
    to examine difficult cases.

14
Cascade of Classifier
  • INTRODUCTION
  • FEATURES
  • Adaptive Boosting
  • Cascade of Classifier
  • RESULTS

15
Cascade of Classifier
  • INTRODUCTION
  • FEATURES
  • Adaptive Boosting
  • Cascade of Classifier
  • RESULTS
  • Given a trained cascade of classifiers, the false
    positive rate and detection rate at each stage
    are
  • And 3 parameters should decide in the
    optimization framework
  • the number of classifier stages
  • the number of features, ni , of each stage
  • the threshold of each stage

16
Cascade of Classifier
  • INTRODUCTION
  • FEATURES
  • Adaptive Boosting
  • Cascade of Classifier
  • RESULTS

Algorithm
17
RESULTS
  • INTRODUCTION
  • FEATURES
  • Adaptive Boosting
  • Cascade of Classifier
  • RESULTS

18
RESULTS
  • INTRODUCTION
  • FEATURES
  • Adaptive Boosting
  • Cascade of Classifier
  • RESULTS

19
RESULTS
  • INTRODUCTION
  • FEATURES
  • Adaptive Boosting
  • Cascade of Classifier
  • RESULTS

20
RESULTS
  • INTRODUCTION
  • FEATURES
  • Adaptive Boosting
  • Cascade of Classifier
  • RESULTS
  • FAILURE MODES
  • Informal observation suggests that the face
    detector can detect faces that are tilted up to
    about 15 degrees in plane and about 45 degrees
    out of plane.
  • We have also noticed that harsh backlighting in
    which the faces are very dark while the
    background is relatively light sometimes causes
    failures.
  • Our face detector fails on significantly occluded
    faces.
Write a Comment
User Comments (0)
About PowerShow.com