+ All Categories
Home > Documents > The Viola/Jones Face Detector

The Viola/Jones Face Detector

Date post: 22-Feb-2016
Category:
Upload: cana
View: 72 times
Download: 0 times
Share this document with a friend
Description:
The Viola/Jones Face Detector. Prepared with figures taken from “Robust real-time object detection” CRL 2001/01, February 2001. Three Measures Toward Speeded Up Detection. Integral image: a fast way to compute simple “features” - PowerPoint PPT Presentation
17
The Viola/Jones Face Detector Prepared with figures taken from “Robust real-time object detection” CRL 2001/01, February 2001
Transcript
Page 1: The Viola/Jones Face Detector

The Viola/Jones Face Detector

Prepared with figures taken from“Robust real-time object detection”

CRL 2001/01, February 2001

Page 2: The Viola/Jones Face Detector

Three Measures Toward Speeded Up Detection

• Integral image: a fast way to compute simple “features”

• In Adaboost the weak learner is nothing but a feature selector. The advantage is that if there are N weak learners there are merely N features to compute.

• Cascaded combination of classifiers. Most of true negatives are rejected very fast at the at the first few stages. Can keep high detection rate and low false positive rate.

Page 3: The Viola/Jones Face Detector

Image Features

Rectangle filters

Similar to Haar wavelets

Base resolution is 24-by-24

11 scales, scaling factor of 1.25

45396 features

Page 4: The Viola/Jones Face Detector

Rectangular Features for Face Detection

Forehead, eye features can be captured

Page 5: The Viola/Jones Face Detector

Fast Feature Computation: Integral Image

• Integral image value at a pixel (x, y) is the sum of the pixel values of the original image above and to the left of (x, y), inclusive.

• Integral image can be computed by one pass through the image

Page 6: The Viola/Jones Face Detector

Computing Sum within a Rectangle by Integral Image

• The sum of the pixels within rectangle D can be computed with four array references.

• The value of the integral image at location 1 is the sum of the pixels in rectangle A.

• The value at location 2 is A + B, at location 3 is A + C, and at location 4 is A + B + C + D.

• The sum within D can be computed as 4 + 1 - (2 + 3).

Page 7: The Viola/Jones Face Detector

Constrained Classifier: Feature Selection

• Restrict the weak learner to a single feature

• A weak classifier hj(x) consists of a feature fj, a threshold j, and a parity pj indicating the direction of inequality sign:

• x is a 24-by-24 sub-window of an image

.otherwise,0

,)(if1)( jjjj

j

θpxfpxh

Page 8: The Viola/Jones Face Detector

Boosting Algorithm

Page 9: The Viola/Jones Face Detector

Learning Result

Must do better

Page 10: The Viola/Jones Face Detector

Cascading Classifiers

The initial classifier eliminates a large number of negative examples with very little processing.

Subsequent layers eliminate additional negatives but require additional computation.

After several stages of processing the number of sub-windows have been reduced radically.

Page 11: The Viola/Jones Face Detector

How Cascading Can Meet Performance?

K

iifF

1

For K stages of cascading with each stage having fi as the false positive rate, the overall false positive rate for the cascade is

Similarly, the overall detection rate is

K

iidD

1

To keep F very low and D very high, for each stage the goal is to have very high detection rate (close to 100%),

but moderate false positive rate (say, 30%)

Page 12: The Viola/Jones Face Detector

Cascaded Classifier

1 Feature 5 Features

F

50%20 Features

20% 2%FACE

NON-FACE

F

NON-FACE

F

NON-FACE

IMAGESUB-WINDOW

• A 1 feature classifier achieves 100% detection rate and about 50% false positive rate.

• A 5 feature classifier achieves 100% detection rate and 40% false positive rate (20% cumulative)– using data from previous stage.

• A 20 feature classifier achieve 100% detection rate with 10% false positive rate (2% cumulative)

Page 13: The Viola/Jones Face Detector

Building A Cascaded Detector

Page 14: The Viola/Jones Face Detector

Classifier is Learned from Labeled Data

• Training Data– 4916 hand labeled faces

• All frontal

– 10000 non faces– Faces are normalized

• Scale, translation

• Many variations– Across individuals– Illumination– Pose (rotation both in plane and out)

Page 15: The Viola/Jones Face Detector

Boosted Face Detection

• For each round of boosting:– Evaluate each rectangle filter on each example– Sort examples by filter values– Select best threshold for each filter (min Z)– Select best filter/threshold (= Feature) – Reweight examples

• Weeks to learn train• 15 frames per second to detect faces from unknown

images.

Page 16: The Viola/Jones Face Detector

Performance

Page 17: The Viola/Jones Face Detector

Output of Face Detector on Test Images


Recommended