![Page 1: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/1.jpg)
Soft Biometrics and Continuous AuthenticationDR. TERENCE SIMSCHOOL OF COMPUTINGNATIONAL UNIVERSITY OF SINGAPORE
![Page 2: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/2.jpg)
Brief Bio
• Associate Professor & Vice Dean
• Research: face recognition, biometrics, computational photography
• PhD from CMU, MSc from Stanfrod, SB from MIT
• Google “Terence Sim”, or [email protected]
![Page 3: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/3.jpg)
![Page 4: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/4.jpg)
Traditional authentication: one-time
![Page 5: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/5.jpg)
Session hijacking
System still thinks legitimate user is there!Solution: continuous authentication
![Page 6: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/6.jpg)
Cassandra CarrilloMSc. Thesis 2003
![Page 7: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/7.jpg)
R Janakiraman, S Kumar, S Zhang, T Sim 2005
• Using Continuous Face Verification to Improve Desktop Security
![Page 8: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/8.jpg)
INTRODUCTION
![Page 9: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/9.jpg)
![Page 10: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/10.jpg)
#1: Must be done passively
• Asking for PIN repeatedly causes frustration
• Biometrics is best suited for this
![Page 11: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/11.jpg)
#2: Have minimal overhead
• Usability & energy issues
![Page 12: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/12.jpg)
#3: Achieve low error rates
• High FAR: imposter easily takes over
• High FRR: re-login needed, user is inconvenienced
• Time must be taken into account• FAR & FRR not enough; • new performance metric needed
![Page 13: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/13.jpg)
#4: Provide Authentication Certainty at all times
• Certainty that the legitimate user is still present
• Even when user provides no biometric signals
![Page 14: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/14.jpg)
![Page 15: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/15.jpg)
CRITERIA
![Page 16: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/16.jpg)
Observations over time
![Page 17: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/17.jpg)
#1: Account for reliability of different modalities
• Fingerprint considered more reliable than face
• Thus must affect the authentication decision more than face
![Page 18: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/18.jpg)
#2: Older observations must be discounted to reflect the increasing uncertainty of the continued presence of the legitimate user
• The longer the elapsed time, the more uncertain is the continued presence of the user.
![Page 19: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/19.jpg)
#3: It must be possible to determine authentication certainty at any point in time, even when there is no observations in one or more modalities
• At any time, the system must be able to check if the legitimate user is still present.
![Page 20: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/20.jpg)
CRITERIA
![Page 21: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/21.jpg)
![Page 22: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/22.jpg)
![Page 23: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/23.jpg)
System Architecture
Integrator
DRV
User space
Kernel space User ok/not ok(actually delay jiffies)
callback
If user not ok, freeze/delay process.If user ok, continue
with system call without delay.
system call
P1 P2 P3
KDM+pam
![Page 24: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/24.jpg)
Probabilistic Approach
• The Integrator computes a probabilistic estimate of user presence, Psafe.
• The OS is tuned with a threshold for verification, Tsafe.• If Psafe < Tsafe, then user deemed absent.
• OS processes belonging to the user’s interactivesession are suspended or delayed as a function of
(Psafe- Tsafe, syscall)
![Page 25: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/25.jpg)
Hidden Markov Model
![Page 26: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/26.jpg)
HMM States
Safe
User still present at console.
Attacked
User is absent, orImposter has hijacked console.
1 - p
0
p1
p: prob. of remaining in Safe state at next time instant.
![Page 27: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/27.jpg)
Bayesian Inference• Let zt be a biometric observation (face or fingerprint) at
time t.
• Let xt be the state at time t.
• Given the current and past observations, what is the most likely current state?
• Bayesian inference: select the larger ofP(xt=Safe | z1, z2, … zt ) and
P(xt=Attacked | z1 , z2 , … zt )
![Page 28: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/28.jpg)
Bayesian Inference
• P(xt | z1, …, zt ) is efficiently computed in terms of
• P(zt | xt ) : prob. of getting current observation given current state
• P(xt | xt-1 ) : transition probabilities
• P(xt-1 | z1, …, zt-1 ) : previous state given previous observations (recursion)
• Upon initial login, • t=0, and P(x0=Safe) = 1
![Page 29: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/29.jpg)
Face Biometric
• We use a Bayesian classifier.
• From 500 training face images of legitimate user, and 1200 images of other people (imposter), we learn:
P(y | user) P(y | imposter)
Face feature y
![Page 30: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/30.jpg)
Face Biometric
• Note that • P(zt | xt = Safe) is just P(y | user)• P(zt | xt = Attacked) is just P(y | imposter)
![Page 31: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/31.jpg)
Fingerprint Biometric
• Also Bayesian classifier.
• Vendor’s proprietary algorithm matches 2 fingerprint images.• Outputs a matching score, s
• From training images, we learn:• P(s | user) and P(s | imposter)
• Which become• P(zt | xt = Safe) and P(zt | xt = Attacked) respectively
![Page 32: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/32.jpg)
Further Comments• Psafe = P(xt=Safe | z1, …, zt )
• We can compute Psafe anytime. • If no observation at time t, then use most recent observation:
Psafe = P(xt=Safe | z1, …, zt-1 ) • But decay transition probability p by time lapse.
p = e kΔt
• This reflects increasing uncertainty about presence of user when no observations available.
![Page 33: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/33.jpg)
Further Comments• In theory, we want the larger of
P(xt=Safe | z1,…, zt ) and P(xt=Attacked | z1,…, zt )
• Equivalent to: Psafe > 0.5
• But in practice, we use Psafe > Tsafe• More flexible: different Tsafe for different process actions (e.g.
reads vs. writes)• Avoids “close call” cases when both probabilities almost equal.
• Math details in paper.
![Page 34: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/34.jpg)
![Page 35: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/35.jpg)
![Page 36: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/36.jpg)
Other Fusion Methods
x1 x2 x3 x4
Temporal-first
Psafe
![Page 37: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/37.jpg)
Other Fusion Methods
Psafe
Modality-first
y1
y2
![Page 38: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/38.jpg)
Naïve Integration
• Idea: use the most reliable modality available at any time instant.
• Since fingerprint more reliable than face, use it whenever available.
• Else use face.
• If no modality available, use the previous one, but decay it appropriately.
![Page 39: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/39.jpg)
Reliability
![Page 40: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/40.jpg)
Experiment: Legitimate User
• Indiv. Probabilities sporadic significant FAR/FRR for any threshold Tsafe
• FAR = security breach!
• FRR = inconvenience
• Holistic Fusion closest to ideal.
• Abrupt drop in Temporal-first, Modality-first curves.
![Page 41: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/41.jpg)
Experiment:Imposter
• Imposter hijacks session at time = 38s
• Detect by change in slope.
• Holistic Fusion and Naïve Integration detects hijacking sooner than others (time = 43s).
![Page 42: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/42.jpg)
Experiment:Partial Impersonation
• Successfully faked fingerprint, but not face.
• This is easily detected by Holistic and Naïve, but not by others.
![Page 43: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/43.jpg)
Psafe for different tasks
![Page 44: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/44.jpg)
Usability test
• 58 people to perform different tasks
![Page 45: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/45.jpg)
Usability test
• CBAS verifies users at a low FRR, and low FAR.
• Surprising result: (a) no statistical evidence to show that CBASoverhead affects task efficiency; (b) system performancedegradation was imperceptible by users.
• Many users felt uncomfortable being “watched” by webcam.Discreet placement may solve this.
• A biometric solution for continuous authentication is practicaland usable.
• Multi-core processors will further reduce the overhead.
![Page 46: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/46.jpg)
New Performance Metric
• Time to Correct Reject (TCR)
• The interval between the start of the first action taken by the imposter to the time instant that the system decides to (correctly) reject him.
• Ideally, TCR = 0.• Practically, TCR < W (minimum time for the imposter to
damage the system, eg. To type “rm –rf *”)• As long as TCR < W, system integrity is assured
![Page 47: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/47.jpg)
New Performance Metric
• Probability of Time to Correct Reject (PTCR)
• The probability that TCR is less than W
• Ideally, PTCR = 1.• Practically, PTCR < 1 may be tolerable• This means that sometimes, the system can take longer
than W seconds to correctly reject an imposter.• If system always fails to correctly reject, then PTCR = 0
for all W• PTCR is analogous to FAR
![Page 48: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/48.jpg)
New Performance Metric
• Usability
• the fraction of the total time that the user is granted access to the protected resource• eg. User logs in for a total duration of T, but system
sometimes rejects user• Let t be the total time user is accepted• Then Usability = t / T
• Ideally, Usability = 1.• Usability is analogous to FRR
![Page 49: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/49.jpg)
New Performance Metric
• Usability-Security Characteristic Curve (USC)
• Plot of Usability vs PTCR
• Analogous to ROC curve
![Page 50: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/50.jpg)
USC curve for our system
![Page 51: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/51.jpg)
![Page 52: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/52.jpg)
![Page 53: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/53.jpg)
![Page 54: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/54.jpg)
Soft biometrics: Definition
• those characteristics that provide some information about the individual, but lack the distinctiveness and permanence to sufficiently differentiate any two individuals under normal circumstance• e.g. gender, clothes color
![Page 55: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/55.jpg)
System
• Hard biometric: face recognition (eigenface)
• Soft biometric: face color histogram, clothes colorhistogram
![Page 56: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/56.jpg)
4 modes
![Page 57: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/57.jpg)
Hard vs Soft biometrics
![Page 58: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/58.jpg)
Hard vs Soft biometrics
Computational time/ Energy
Accuracy
Face
Clothes color
Iris
Gender
![Page 59: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/59.jpg)
![Page 60: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/60.jpg)
![Page 61: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/61.jpg)
Coping with illum change
![Page 62: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/62.jpg)
Coping with illum change
![Page 63: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/63.jpg)
Evaluation
![Page 64: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/64.jpg)
Evaluation
![Page 65: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/65.jpg)
Evaluation
![Page 66: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/66.jpg)
![Page 67: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/67.jpg)
![Page 68: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/68.jpg)
Smartphones
• New opportunity for Continuous Authentication
• Rich sensors:
![Page 69: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/69.jpg)
Possible biometrics• Face: gender, identity, age, race, expression• Iris?• Voice• Gait• Keystroke dynamics (touch)• Fingerprint• Location• Wifi signature• Cellular signature
![Page 70: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/70.jpg)
Energy usage is critical!
Computational time/ Energy
Accuracy
Face
Clothes color
Iris
Gender
![Page 71: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/71.jpg)
• Most research use touch dynamics
• Multimodal biometrics will be more useful
• Computational efficiency not yet considered
• Possibility for forensics use
![Page 72: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/72.jpg)
![Page 73: Soft Biometrics and Continuous Authentication · • Surprising result: (a) no statistical evidence to show that CBAS overhead affects task efficiency; (b) system performance degradation](https://reader035.vdocuments.us/reader035/viewer/2022070904/5f6fea53417d5254591996a8/html5/thumbnails/73.jpg)
References• Sim, Terence, Sheng Zhang, Rajkumar Janakiraman, and
Sandeep Kumar. "Continuous verification using multimodal biometrics." IEEE transactions on pattern analysis and machine intelligence 29, no. 4 (2007): 687-700.
• Kwang, Geraldine, Roland HC Yap, Terence Sim, and Rajiv Ramnath. "An usability study of continuous biometrics authentication." In International Conference on Biometrics, pp. 828-837. Springer Berlin Heidelberg, 2009.
• Niinuma, Koichiro, Unsang Park, and Anil K. Jain. "Soft biometric traits for continuous user authentication." IEEE Transactions on information forensics and security 5, no. 4 (2010): 771-780.
• Janakiraman, Rajkumar, and Terence Sim. "Keystroke dynamics in a general setting." In International Conference on Biometrics, pp. 584-593. Springer Berlin Heidelberg, 2007.
• Traore, Issa, ed. Continuous Authentication Using Biometrics: Data, Models, and Metrics: Data, Models, and Metrics. IGI Global, 2011.