Presentation is loading. Please wait.

Presentation is loading. Please wait.

A Brief Survey on Face Recognition Systems Amir Omidvarnia March 2007.

Similar presentations


Presentation on theme: "A Brief Survey on Face Recognition Systems Amir Omidvarnia March 2007."— Presentation transcript:

1

2 A Brief Survey on Face Recognition Systems Amir Omidvarnia March 2007

3 Outline  Introduction  Face Recognition Concepts  2D Face recognition Systems  An Example  3D Face Recognition  Suggestions

4 Introduction  What is face recognition?  Applications Security applications Security applications Image search engine Image search engine

5 Requirements  Accurate  Efficient  Light invariant  Rotation invariant

6 Applications of Face Recognition  Desire to locate specific individuals Criminals Criminals TERRORISTS TERRORISTS Missing Children Missing Children Surveillance Surveillance

7 Face Recognition Concepts  Enrollment An initial feature set is constructed from the relevant physical traits of the user.

8 Face Recognition Concepts  Verification Extracted feature set from each person is compared with the enrollment feature set. If the resulting score value is above a predefined threshold, the user is considered to be authenticated.

9 Face Recognition Concepts  Identification In contrast to the verification use case, with identification the (claimed) identity of the user is not known in advance, but shall be determined based on sample images of the user's face and a set (population) of feature sets with known identities.

10 An Example: FaceVACS Architecture  Face Localization  Eye Localization  Image Quality Check  Normalization  Preprocessing  Feature Extraction  Construction of the Reference Set  Comparison

11 Face Recognition Concepts  The Facial Identification Record (FIR) In the result of processing the raw samples (images), e.g. during enrollment, feature sets are created. In the context of FaceVACS-SDK we use the term FIR for these feature sets.

12 Face Recognition Concepts  Enrollment and Verification

13 Face Recognition Concepts  FAR, FRR and EER FAR (False Acceptance Rate) is the probability that a sample falsely matches the presented FIR. FRR (False Rejection Rate) is the probability that a sample of the right person is falsely rejected. The value of FAR and FRR at the point where the plots cross is called the Equal Error Rate (EER).

14 Face Recognition Concepts  FAR, FRR and EER

15 Face Recognition Systems  Feature-Based  Appearance-Based  Model-Based

16 Feature-Based Algorithms  Geometric Features  Texture  Skin color  Multiple features

17 Appearance-Based Algorithms  Eigenface  Fisherface  SVM  Neural Networks  Hidden Markov Models

18 Model-Based Algorithms  Face Bunch Graph  Predefined face templates  Deformable templates

19 An Example  Sample Image

20 An Example  Eye location found by the algorithm

21 An Example  After Normalization

22 An Example  After Preprocessing

23 An Example  Extracting local features

24 An Example  Forming the reference set of the image

25 Image Database  Effective Factors in combining FIRs  Influence and arrangement of lighting conditions  Sample Quality  Orientation of Samples  Adornment  Face Angle  Face Appearance

26 3D Face Recognition

27

28 Suggestions  A comprehensive Survey on 2D Face Recognition Algorithms Face Detection Face Detection Face Segmentation Face Segmentation Feature Extraction Feature Extraction Facial Models Facial Models Texture Analysis Texture Analysis  Towards 3D Face Recognition  Combining other Biometrics such as Iris Recognition  Towards Multimodal Systems


Download ppt "A Brief Survey on Face Recognition Systems Amir Omidvarnia March 2007."

Similar presentations


Ads by Google