Can Facial Uniqueness be Inferred from Impostor Scores? Abhishek - - PowerPoint PPT Presentation
Can Facial Uniqueness be Inferred from Impostor Scores? Abhishek Dutta Oct 14, 2013, Nijmegen, Netherlands. Presentation accompanying the following paper: A. Dutta, R. Veldhuis and L. Spreeuwers. Can Facial Uniqueness be Inferred from Impostor
Abhishek Dutta
Oct 14, 2013, Nijmegen, Netherlands. Presentation accompanying the following paper:
Nijmegen, Netherlands.
Image Pair 1
Image Pair 2
Image Pair 1 Image Pair 2
person-id 261 person-id 263 person-id 272
Irrespective of your decision {yes, no, maybe} , you were:
◮ more uncertain in your decision for Image Pair 1, ◮ less uncertain in your decision for Image Pair 2.
Image Pair 1 Image Pair 2
person-id 261 person-id 263 person-id 272
If we assume that faces form a Gaussian cloud in a high dimensional space, then
◮ identity relates to direction, and ◮ distinctness relates to distance from the mean.
Therefore, we are more confident in making decision about identity when facial images are well separated in the face space – by virtue
◮ Therefore, in Biometrics, researchers have been actively
involved in measuring uniqueness from facial photographs
◮ Such facial uniqueness measurements are useful to build an
face recognition system that adaptively vary the decision threshold to improve recognition performance.
Impostor Population
how unique is this facial image? with respect to this impostor population? similarity score count count similarity score fairly common face highly distinct face
Impostor Population
how unique is this facial image? with respect to this impostor population? similarity score count count similarity score fairly common face highly distinct face
Assumption
◮ similarity score is influenced only by facial identity ◮ non-unique facial image1 will generally exhibit high level of
similarity to many other subjects in a large population (by definition) Hypothesis facial uniqueness of a subject can be inferred from its impostor similarity score distribution
1lamb in the biometric zoo
In this paper, we investigate the stability of facial uniqueness measures based on impostor scores. For this, we perform following two experiments:
Experiment 1
To study the influence of image quality variations on the impostor score distribution.
Experiment 2
To study the stability of a recently proposed Impostor Based Uniqueness Measure (IUM) of [Klare and Jain 2013] under image quality variations.
Probe (Average Face) Baseline Gallery Motion Blur (angle = 0)
Gaussian Noise (mean= 0)
Pose (camera-id)
05_1 04_1 19_0 14_0 13_0 08_0 05_0 19_1 08_1
Gallery with Image Quality Variation Baseline Impostor Score Distribution Quality based Impostor Score Distribution