detection on faces

22
Detection on Faces Detection on Faces Levente Sajó

Upload: hoang

Post on 22-Feb-2016

25 views

Category:

Documents


0 download

DESCRIPTION

Detection on Faces. Levente Saj ó. Detecting Face. Knowledge-Based Methods A face often appears in an image with two eyes that are symmetric to each other, a nose and a mouth Problem: it is difficult to translate human knowledge into well defined rules Feature-Invariant Approaches - PowerPoint PPT Presentation

TRANSCRIPT

Page 1: Detection on Faces

Detection on FacesDetection on Faces

Levente Sajó

Page 2: Detection on Faces

2008.02.28

2

Detecting Faces

Boosted Cascade Detector

Facial Expression Recognition

Shape Models

Localizing Facial Features

Next Steps...

Detecting FaceDetecting Face

• Knowledge-Based Methods– A face often appears in an image with two

eyes that are symmetric to each other, a nose and a mouth

– Problem: it is difficult to translate human knowledge into well defined rules

• Feature-Invariant Approaches– Facial Features

• Using edge detectors, facial features such as eyes, nose, mouth, and hair line are extracted

• A statistical model is built to describe their relationships

• Problem: lighting, shadows– Skin Color

Page 3: Detection on Faces

2008.02.28

3

Detecting Faces

Boosted Cascade Detector

Facial Expression Recognition

Shape Models

Localizing Facial Features

Next Steps...

Detecting FaceDetecting Face

• Template Matching Methods– Predefined templates

• correlation values with the standard patterns are computed eg.: for the face contour, eyes, nose and mouth independently

• it can not effectively deal with variation in scale, pose and shape

– Deformable Templates – Shape Models

• How to represent the template?

• How to model deformations?• Efficient matching

algorithms

Page 4: Detection on Faces

2008.02.28

4

Detecting Faces

Boosted Cascade Detector

Facial Expression Recognition

Shape Models

Localizing Facial Features

Next Steps...

Detecting FaceDetecting Face

Appearance-Based Methods– Eigenfaces – PCA– Neural Networks– SVM– AdaBOOST (by Viola and Jones, 2001)

• Quite accurate• Fastest – 10-15-times faster then the

others• An extended version is implemented in

opencv (by Lienhart, 2002)• This method is used in Google Portrait,

too

Page 5: Detection on Faces

2008.02.28

5

Detecting Faces

Boosted Cascade Detector

Facial Expression Recognition

Shape Models

Localizing Facial Features

Next Steps...

Boosted Cascade DetectorBoosted Cascade Detector

• Consists 3 parts– Encoding the image data known as

“integral image”– Running the boosting algorithm

known as AdaBoost, to select the appropriate feature

– a cascade of templates that allows simple feature sets to quickly discard most of the uninteresting parts of the image

Page 6: Detection on Faces

2008.02.28

6

Detecting Faces

Boosted Cascade Detector

Facial Expression Recognition

Shape Models

Localizing Facial Features

Next Steps...

Integral ImageIntegral Image• An integral image is constructed by replacing

each image pixel value i(x,y) with a value that corresponds to the pixel sum, above and to the left of the pixel

Page 7: Detection on Faces

2008.02.28

7

Detecting Faces

Boosted Cascade Detector

Facial Expression Recognition

Shape Models

Localizing Facial Features

Next Steps...

Feature ClassifiersFeature Classifiers• Each classifier consists of a set of adjacent positive and

negative regions• Different feature types:

• If template is 24x24, then it produces 30000 different feature classifiers

• Each individual feature classifier attempts to discriminate between faces/non-faces by summing the difference between adjacent regions

Page 8: Detection on Faces

2008.02.28

8

Detecting Faces

Boosted Cascade Detector

Facial Expression Recognition

Shape Models

Localizing Facial Features

Next Steps...

Boosting AlgorithmBoosting Algorithm• Boosting algorithms aim to combine

many “weak classifiers” into one effective “strong classifier”.

• AdaBoost achieves this by selecting the more promising classifiers.

Page 9: Detection on Faces

2008.02.28

9

Detecting Faces

Boosted Cascade Detector

Facial Expression Recognition

Shape Models

Localizing Facial Features

Next Steps...

Cascaded DetectorsCascaded Detectors• A series of classifiers are applied to every sub-window. • The initial classifier eliminates a large number of

negative examples with very little processing. • Subsequent layers eliminate additional negatives

Page 10: Detection on Faces

2008.02.28

10

Detecting Faces

Boosted Cascade Detector

Facial Expression Recognition

Shape Models

Localizing Facial Features

Next Steps...

BCD ExperimentsBCD Experiments

• Detecting whole face• Detecting facial features

• The whole face template is more robust than any of the facial feature templates

Page 11: Detection on Faces

2008.02.28

11

Detecting Faces

Boosted Cascade Detector

Facial Expression Recognition

Shape Models

Localizing Facial Features

Next Steps...

Facial Expression RecognitionFacial Expression Recognition

• Recognizing the basic emotional expressions from video– neutral, anger, joy, sadness, surprise,

disgust, fear• Steps:

– Localizing the face– Preprocessing !!– Classification (with SVM, AdaBoost)

Page 12: Detection on Faces

2008.02.28

12

Detecting Faces

Boosted Cascade Detector

Facial Expression Recognition

Shape Models

Localizing Facial Features

Next Steps...

Preprocessing 1Preprocessing 1

• Gábor Transformation (Bartlett, 2003)– 48x48 face images were used– Normalized so, that the approximate distance

between center of eyes to be 24 pixel– The images were converted into a Gabor

magnitude representation, using a bank of Gabor filters at 8 orientations and 5 spatial frequencies

Page 13: Detection on Faces

2008.02.28

13

Detecting Faces

Boosted Cascade Detector

Facial Expression Recognition

Shape Models

Localizing Facial Features

Next Steps...

Preprocessing 2Preprocessing 2• Feature Extraction (Michel, El Kaliouby, 2003)

– In earlier work, using an SVM classifier on displacements of manually defined facial landmarks in image sequences yielded high classification accuracy

– Uses a face template to initially locate the position of the 22 facial features

– Their position is tracked over subsequent frames– A vector of feature displacements is calculated by

taking the Euclidean distance between feature locations in a neutral and a “peak” frame

Page 14: Detection on Faces

2008.02.28

14

Detecting Faces

Boosted Cascade Detector

Facial Expression Recognition

Shape Models

Localizing Facial Features

Next Steps...

Shape ModelsShape Models

• Predefined Shape Models– Setting the position of the landmarks

manually• Active Shape Models (Cootes, 1995)

– Using a set of manually labeled face image as a training set

– PCA is used to reduce the dimensionality of the training set

– During the training stage, for each landmarks their mean position and variance are determined

Page 15: Detection on Faces

2008.02.28

15

Detecting Faces

Boosted Cascade Detector

Facial Expression Recognition

Shape Models

Localizing Facial Features

Next Steps...

Shape ModelsShape Models

Page 16: Detection on Faces

2008.02.28

16

Detecting Faces

Boosted Cascade Detector

Facial Expression Recognition

Shape Models

Localizing Facial Features

Next Steps...

Localizing Facial FeaturesLocalizing Facial Features• Using face detection

techniques to detect facial features – local feature detectors (Cristinacce, Cootes, 2004)– Normalised correlation– BCD

• Shape models can be used to reduce the number of false detections by only selecting plausible configurations of feature matches

• With shape models the approximate location and orientation of the face can be used to perform a local search of facial features.

Page 17: Detection on Faces

2008.02.28

17

Detecting Faces

Boosted Cascade Detector

Facial Expression Recognition

Shape Models

Localizing Facial Features

Next Steps...

Shape SearchShape Search

• Unconstrained search– A bounding box of every feature are

calculated– The feature detector is then applied

to each region– The best match of is detector is taken

• Constrained search – Combinatoric Shape Search (CSS)– Few of best responses are taken from

each local detector– All combinations of feature locations

are checked

Page 18: Detection on Faces

2008.02.28

18

Detecting Faces

Boosted Cascade Detector

Facial Expression Recognition

Shape Models

Localizing Facial Features

Next Steps...

Next Steps…Next Steps…• Previous method works fine on neutral

faces• The same method can be modified to

work on faces with emotions• Other approach for localizing facial

features?– Color based– Intensity based (edges, thresholds)

Page 19: Detection on Faces

2008.02.28

19

Detecting Faces

Boosted Cascade Detector

Facial Expression Recognition

Shape Models

Localizing Facial Features

Next Steps...

Next Steps…Next Steps…

Page 20: Detection on Faces

2008.02.28

20

Detecting Faces

Boosted Cascade Detector

Facial Expression Recognition

Shape Models

Localizing Facial Features

Next Steps...

Next Steps…Next Steps…

Page 21: Detection on Faces

2008.02.28

21

Detecting Faces

Boosted Cascade Detector

Facial Expression Recognition

Shape Models

Localizing Facial Features

Next Steps...

Next Steps…Next Steps…

Page 22: Detection on Faces

2008.02.28

22

Detecting Faces

Boosted Cascade Detector

Facial Expression Recognition

Shape Models

Localizing Facial Features

Next Steps...

Next Steps…Next Steps…