1 / 9

Eye Tracking Using A Simple Webcamera

Eye Tracking Using A Simple Webcamera. David Wild Supervisor: James Connan Rhodes University Computer Science Department. Contents:. Brief Recap of project Approaches found My choice Tools Still Needs to be done Questions. Brief Recap of project. Eye tracking:

ania
Télécharger la présentation

Eye Tracking Using A Simple Webcamera

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Eye Tracking Using A Simple Webcamera David Wild Supervisor: James Connan Rhodes University Computer Science Department

  2. Contents: • Brief Recap of project • Approaches found • My choice • Tools • Still Needs to be done • Questions

  3. Brief Recap of project • Eye tracking: • Find the eyes in a frame • Use the eyes to calculate point of interest • Usually pre-program calibration to find the limits • Original Plan • Use an infrared camera to make the detection simpler, still a possibility • Find, or make an appropriate algorithm to calculate the gaze

  4. Approaches found (1) • Real Time Tracking Algorithm with head movement Compensation (RTAC) • Eye detection using the corneal glint • Find the glint-pupil vector • Map to 3D • Calculate head movement • Compensate for head movement

  5. Approaches found (2) • Neural Networks • Find the eye and face using intensities of pixels • Use OpenCv to detect the eye and face regions • Create a smaller picture for training the NN • Use the intensities of each pixel as a input for the NN • Starburst • Noise reduction (shot and line) • Corneal reflection detection using adaptive brightness-threshold • Feature Detection • Ellipse Detection • Calibration

  6. My Choice • Starburst Eye detection • Robust system • Interesting to examine • Could possibly include head movement compensation as a extension to the original algorithm • Can still change this choice

  7. Tools • C++ • OpenCV • Ordinary USB Webcam

  8. Still needs to be done • Obviously Coding and Implementation • Finding the eye • Design the Calibration pre-program • Implement the algorithm • Test and improve

  9. Questions?

More Related