computer vision: eye tracking by: geraud campion michael o’connor
TRANSCRIPT
![Page 1: Computer Vision: Eye Tracking By: Geraud Campion Michael O’Connor](https://reader036.vdocuments.us/reader036/viewer/2022082712/56649f345503460f94c50cfa/html5/thumbnails/1.jpg)
Computer Vision:Eye Tracking
By:
Geraud Campion
Michael O’Connor
![Page 2: Computer Vision: Eye Tracking By: Geraud Campion Michael O’Connor](https://reader036.vdocuments.us/reader036/viewer/2022082712/56649f345503460f94c50cfa/html5/thumbnails/2.jpg)
Frame Work
A brief overview of eye tracking, formulas for image dissection, and some current applications of eye tracking.
![Page 3: Computer Vision: Eye Tracking By: Geraud Campion Michael O’Connor](https://reader036.vdocuments.us/reader036/viewer/2022082712/56649f345503460f94c50cfa/html5/thumbnails/3.jpg)
Eye Tracking
Why Eyes? Failures of facial
recognition due to poor alignment
Eye and eye movement are important to human interaction
![Page 4: Computer Vision: Eye Tracking By: Geraud Campion Michael O’Connor](https://reader036.vdocuments.us/reader036/viewer/2022082712/56649f345503460f94c50cfa/html5/thumbnails/4.jpg)
Eye Tracking
Current Approaches Visible Spectrum Cameras Near-Infra-Red cameras (NIRs)
Work well in optimal conditions: fast and accurate
Not so good otherwise: a lot of false positives Not a great help in the field of psychology or
neurology
![Page 5: Computer Vision: Eye Tracking By: Geraud Campion Michael O’Connor](https://reader036.vdocuments.us/reader036/viewer/2022082712/56649f345503460f94c50cfa/html5/thumbnails/5.jpg)
Another Technique
Reflected Light from the Eye:
![Page 6: Computer Vision: Eye Tracking By: Geraud Campion Michael O’Connor](https://reader036.vdocuments.us/reader036/viewer/2022082712/56649f345503460f94c50cfa/html5/thumbnails/6.jpg)
Reflected Light
![Page 7: Computer Vision: Eye Tracking By: Geraud Campion Michael O’Connor](https://reader036.vdocuments.us/reader036/viewer/2022082712/56649f345503460f94c50cfa/html5/thumbnails/7.jpg)
Eye Tracking
Search Methods Probabilistic Methods
Bayesian Inference Model
Key to this is that an image is cut into a collage of rectangles of arbitrary size
![Page 8: Computer Vision: Eye Tracking By: Geraud Campion Michael O’Connor](https://reader036.vdocuments.us/reader036/viewer/2022082712/56649f345503460f94c50cfa/html5/thumbnails/8.jpg)
Eye Tracking
Y is a random matrix y is a specific point A = {a1, a2, … an} where ai is a
rectangle in Y H = {H1, H2, … Hn} is a random vector
assigning each patch Hi a value: 1 object of interest, -1 background, 0 not rendered
![Page 9: Computer Vision: Eye Tracking By: Geraud Campion Michael O’Connor](https://reader036.vdocuments.us/reader036/viewer/2022082712/56649f345503460f94c50cfa/html5/thumbnails/9.jpg)
Eye Tracking
All this leads to this formula P(H = 1 | y) =
Σ [P(H=1) p(h | Hi = 1)p(y | hiHi = 1) ]/ p(y)
which is the probability that a portion y holds our object of interest:
![Page 10: Computer Vision: Eye Tracking By: Geraud Campion Michael O’Connor](https://reader036.vdocuments.us/reader036/viewer/2022082712/56649f345503460f94c50cfa/html5/thumbnails/10.jpg)
Eye Tracking
Situation Based Reference Make a hierarchy of “context dependent
experts” Each expert uses probabilistic methods Then we use this formula:
p(o|y) = ∫p(s|y) p(o|sy) dh Y – an observed image S – contextual situation O – location of left eye of the face on image
![Page 11: Computer Vision: Eye Tracking By: Geraud Campion Michael O’Connor](https://reader036.vdocuments.us/reader036/viewer/2022082712/56649f345503460f94c50cfa/html5/thumbnails/11.jpg)
Applications
Camera Mouse Eyebrow/blink patterns for clicking
Driver Fatigue Detection (750 deaths, 20,000 injuries / yr from
commercial vehicles) Detecting Amblyopia in Children Toys
![Page 12: Computer Vision: Eye Tracking By: Geraud Campion Michael O’Connor](https://reader036.vdocuments.us/reader036/viewer/2022082712/56649f345503460f94c50cfa/html5/thumbnails/12.jpg)
Some Results
One Person Two People Eyebrow Clicker VTOY