Idea Details

Team and Contact Details

SNo Student Name School Degree Year Email
1Bilal MusaniSEECSUndergraduateFourth14bscsbmusani@seecs.edu.pk
1Shaheer HashimSEECSUndergraduateFourth14bscsshashim@seecs.edu.pk
1Muhammad Umair KhanSEECSUndergraduateFourth14bscsmumair@seecs.edu.pk

Inter School Idea ? No
Do you need expertises from another area: No
If Yes please provide details of expertises you need:

Idea Details

Idea Name: 3D-FREDI
Slogan: Facial Recognition
Supervisor Name: Dr Muhammad Shahzad
Supervisor Designation: Assistant Professor
Supervisor School: SEECS
Supervisor Department: Department of Computing
Contact number: +923335806389
Email ID: muhammad.shehzad at seecs.edu.pk
Abstract:
    Our project aimed at creating a realistic 3D construct of an individual using readily available interfaces while at the same time detecting and interpreting the emotions of the said individual.
What is the unmet need in society that your idea will fulfill ?
    Motion and expression capture requires state of the art software and hardware to effectively implement. It goes without saying that such equipment does not come cheap. Our idea attempts to reduce the entry barrier for animators and designers who want to create realistic 3D models within a budget.
Who needs it ? How many would benefit ?
   The aforementioned designers and animators along with any video game programmer who wishes to create realistic 3D models without using any expensive and cumbersome equipment.
How will the solution works
    This project is aimed at creating a 3D facial construct of an individual using a monocular camera. The input image will be modulated via machine learning algorithms, which will attempt to map feature points in an individual’s face onto a generic mask. This modified mask will in turn be fine tuned to match the individual concerned. Additionally, this mask will be generated in real time, to accommodate changes in an individual's expressions and henceforth interpret his/hers emotional state.
Who are your competitors ? How is your solution different
    The students from the University of Nottingham have created a similar product in their paper titled "Large Pose 3D Face Reconstruction from a Single Image via Direct Volumetric CNN Regression". Our approach however utilizes deconvolutional networks and also maps the emotional state of an individual.
Status: new
Entry Date & Time: 2017-12-15 (2240)