Smart Page Turner

Automate eye gaze tracker for assistive reading with OpenCV. Check out demo @ https://youtu.be/o7uCJZYBTVk

Gaze Detection

Dependencies needed for the gaze detection:

• OpenCV
• dlib
• numpy
• imutils
• scipy
• shape_predictor_68_face_landmarks.dat

To run the python module:

python landmark_with_center.py --shape-predictor shape_predictor_68_face_landmarks.dat --camera 0

*if you want to use external webcam, change camera param to 1

To change threshold for detection:

FRAME_THRESHOLD = 10
RATIO_THRESHOLD_LB = 0.45
RATIO_THRESHOLD_UB = 0.55

Web host

Web page for simple proof of concept of page turning.

To run the web host, first run

ifconfig   (LINUX)
ipconfig   (WINDOWS)

to get your ip address and swap it within index.html file

Then run

npm install
node app.js