Facial Recognition Viseme Creation - ON-216

Preferred Disciplines: Computer Science (Master’s level)
Company: Anonymous
Project Length: 4-6 months
Desired start date: As soon as possible
Location: Toronto, ON
No. of Positions: 1
Preferences: UofT; UWaterloo

About the Company: 

A small Toronto software company that has already developed a multiuser, Linux-based speaking viseme based messaging system relevant mobile client apps (IOS&Android)

Project Description:

  • Locate the landmark features of a face and then take subsets of these to define our visemes.
  • R&D:  MAP specific landmark coordinates(e.g. outline of mouth, eyes) of a camera captured face onto a sequence of viseme-phonemes(for English and to reflect 4,5 emotions), to distort the input face by the mapping, and we would save each facial distortion as a viseme for that face. 
  • Use Available Face Recognition Libraries(Python compatible)

Research Objectives:

To develop the described utility as a fast, callable library on a Ubuntu server

Methodology:

Designing an algorithm to extract/create visemes from an image of a human face; using available libraries for facial recognition/landmarks for this purpose

Expertise and Skills Needed:

  • Python programming;
  • Linux Ubuntu;
  • Facial recognition algorithms

For more info or to apply to this applied research position, please

  1. Check your eligibility and find more information about open projects
  2. Interested students need to get the approval from their supervisor and send their CV along with a link to their supervisor’s university webpage by applying through the webform.
Program: