Novel Biometrics at Southampton

1
Novel Biometrics at Southampton
Mark S. Nixon, School of Electronics and Computer
Science, University of Southampton
2
Why Gait?
As a biometric, gait is available at a distance
and difficult to hide
Discovery, 2000
3
Biometrics and Gait

• Emerging gait research
• Gait is non-contact and uses sequences
• Advantages perceivable at distance and hard to
  disguise
• Potential applications security/ surveillance,
  immigration, forensic, medicine?
• Other applications moving objects
• Related fields animation, tracking

Clinical Gait Analysis
MIE Medical Research
Nixon et al, in Jain et al Eds. Biometrics, 1999
4
Where are we now?

• Large and varied databases many techniques
• Recognition rates encouraging
• Emergent work on fusion and wider use

Nixon et al. ESTR 2002 and Sensor Review 2003
5
Gait Recognition Other Analyses

• Data and techniques from CMU, MIT, Georgia and
  Maryland
• Mainly holistic measures
• Encouraging results

6
Gait Extraction and Description

• Modelling Movement
• Pendular thigh motion model
• Coupled and forced oscillator
• Anatomically-guided skeleton

• Silhouette descriptions
• Established statistical analysis
• (Temporal) symmetry
• (Velocity) moments
• (Dynamics of) area

Nixon Carter Shutler, and Grant, ESTR 2002
7
Velocity Moments

• Extension of spatial moments
• Applied to silhouettes
• Selected by ANOVA

Shutler, and Nixon BMVC 2001
8
Other Biometrics? Ear Recognition
Why ears? Ears dont grow old (just longer) and
dont smile!
Image
Transformation
Description
Hurley, Nixon and Carter, Image and Vision
Computing 2002
9
Where will we be? Ear Recognition 2
Chang, Bowyer, Sarkar, Victor, IEEE TPAMI, 2003
10
Where will we be? Data Fusion.
Face 70.5 Ear 71.6 FaceEar 90.9
Combination is statistically significantly better!
Chang, Bowyer, Sarkar, Victor, IEEE TPAMI 2003
11
Other Interests

• Computer vision and image processing
• Upcoming talks
• IEEE Face and Gesture 04 EUSIPCO 04
• Medical image application
• Remotely-sensed image analysis
• Etc