for Authentication in Virtual Reality Head- mounted Display Shiqing - - PowerPoint PPT Presentation

▶
for authentication in virtual reality head
SMART_READER_LITE
LIVE PREVIEW

for Authentication in Virtual Reality Head- mounted Display Shiqing - - PowerPoint PPT Presentation

Mar 29, 2024 182 likes •458 views

OcuLock: Exploring Human Visual System for Authentication in Virtual Reality Head- mounted Display Shiqing Luo , Anh Nguyen , Chen Song, Feng Lin, Wenyao Xu , Zhisheng Yan Georgia State University San Diego State

slide-1
SLIDE 1

OcuLock: Exploring Human Visual System for Authentication in Virtual Reality Head- mounted Display

Shiqing Luo∗, Anh Nguyen∗, Chen Song†, Feng Lin‡, Wenyao Xu§, Zhisheng Yan∗ ∗Georgia State University †San Diego State University ‡Zhejiang University §SUNY Buffalo

slide-2
SLIDE 2

Virtual Reality (VR) technology is boosting.

  • The market size reached 3.6 billion dollars in 2018*.

2

*Viar360, “Virtual reality market size in 2018 with forecast for 2019,” 2019.

slide-3
SLIDE 3

Diverse Applications

3

Entertainment Healthcare Military

slide-4
SLIDE 4

Diverse Applications

4

Entertainment Healthcare Military

slide-5
SLIDE 5

Authentication System

  • Protect HMD from unauthorized access.

5

slide-6
SLIDE 6

State-of-the-art Methods

6

Password Unlock pattern Head motion Body motion

slide-7
SLIDE 7

State-of-the-art Methods

  • Expose authentication actions *.
  • Behaviors change over time.

7

Password Unlock pattern Head motion Body motion

*Ling, Zhen, Zupei Li, Chen Chen, Junzhou Luo, Wei Yu, and Xinwen Fu. "I Know What You Enter on Gear VR." In 2019 IEEE Conference on Communications and Network Security (CNS), pp. 241-249. IEEE, 2019.

slide-8
SLIDE 8

Solution: Human Visual System (HVS) auth.

  • An unobservable solution.
  • Behavioral and physiological biometric.

8

slide-9
SLIDE 9

Challenge 1: HVS Hard to Measure

  • HVS components are hard to measure in VR HMD.
  • Limited space.
  • Dark environment.

9

slide-10
SLIDE 10

Challenges 2: Redundant Training

  • Each new user requires a new classifier.

10

Sample1 Sample2 … Sample10 Sample11 Sample12 … Sample20 Sample21 Sample22 … Sample30 …

classifier1 classifier2 classifier3 … user1 user2 user3 …

slide-11
SLIDE 11

System Architecture

  • Module 1: capture the electrical signals from HVS.
  • Module 2: authenticate EOG samples based on similarity.

11

slide-12
SLIDE 12

Module 1 - Visual Stimuli

12

Fixed-Route (FR) City-Street (CS) Illusion (IL) Eye rotation, blinks Scan path Micro-saccades

slide-13
SLIDE 13

Module 1 - EOG Signal Acquisition

  • Remove interference using filters.

13

slide-14
SLIDE 14

Module 2 - Signal Processing

  • Recognize saccades (S), fixations(F) and blinks(B).
  • Continuous wavelet transform algorithm*.

14

*A. Bulling, J. A. Ward, H. Gellersen, and G. Troster, “Eye movement analysis for activity recognition using electrooculography,” IEEE transactions on pattern analysis and machine intelligence, 2010.

slide-15
SLIDE 15

Module 2 - Authentication

  • Extracts behavioral and physiological features from the EOG signal.

15

Behavioral features: Saccade: duration, start time, location. Fixation: duration, start time, centroid. Physiological features: Eyelid: close speed, open speed, stretch extent. Metabolism intensity. Rotation extent: right, left, up, down.

…

Saccade duration Saccade start

slide-16
SLIDE 16

Module 2 - Authentication

  • Compare sample A and B with template sample T.

16

Comparison algorithm

slide-17
SLIDE 17

Module 2 - Authentication

  • Are A and B the same as template?

17

Classifier

Access granted Access denied

No need to re-train the classifier.

slide-18
SLIDE 18

Experiment - Impersonation Attack

  • 70 participants.
  • Each provides 10 records.
  • Records are partitioned into training and testing sets.
  • 1:1, by subject.
  • In each set, 61075 comparison results (1575 positive, 59500 negative).

18

slide-19
SLIDE 19

Experiment - Impersonation Attack

  • F1 scores of all combinations of matching algorithm and classifiers.

19

FR CS IL

Matching algorithms: Ansari-Bradley test (AB); Mann-Whitney u-test (MW); Two-sample Kolmogorov-Smirnov test (KS); Two-sample Cramer-von Mises test (CM); Two-sample t-test (TS).

slide-20
SLIDE 20

Experiment - Impersonation Attack

  • Best F1 score using AB Test and SVM (linear).

20

FR CS IL

Matching algorithms: Ansari-Bradley test (AB); Mann-Whitney u-test (MW); Two-sample Kolmogorov-Smirnov test (KS); Two-sample Cramer-von Mises test (CM); Two-sample t-test (TS).

slide-21
SLIDE 21

Experiment - Impersonation Attack

  • Low equal error rate: EER(FR)=5.27%; EER(CS)=7.32%; EER(IL)=3.55%.

21

Receiver Operating Characteristic (ROC) Equal Error Rate (EER)

slide-22
SLIDE 22

Experiment - Statistical Attack

  • The attacker calculates the PDF of features from users, then uses the

most probable feature values to generate the forgery.

22

…

slide-23
SLIDE 23

Experiment - Statistical Attack

  • Low impact at equal error rate: EER(FR)=6.93%; EER(CS)=7.93%;

EER(IL)=4.97%.

23

ROC EER

slide-24
SLIDE 24

Experiment - Time Efficiency

  • Trade-off between security and convenience.

24

slide-25
SLIDE 25

Experiment - Temporal Stability

  • 5 participants.
  • The accuracy is stable.

25

slide-26
SLIDE 26

Conclusion

  • We propose an EOG-based framework to measure the HVS as a whole

for VR authentication.

  • We design a record-comparison driven authentication scheme.
  • We perform an extensive evaluation of the proposed OcuLock system.

26

Thank you