Deep Neural Networks for Local Editing of Facial Polygon Meshes

Artists in the industry are eager for tools to simplify the modeling of faces for 3D characters. For a while, researchers, including ourselves, used blend shape style models for reconstructing and modeling faces. More recently, researchers investigated using Artificial Intelligence (AI) models, again mostly for face reconstruction from images. AI is more able to properly model faces than blend shape models, yet using AI for modeling is an area that has not yet received much attention. We will design and tune our AI model using two face data sets (one public and the other from our industrial partner). We will compare our new AI model to our previous model, by reproducing functionalities already found in our previous model. We will also compare our new results to those of our previous model to validate the performance of the new AI model.

Faculty Supervisor:

Eric Paquette

Student:

Partner:

Ubisoft Toronto

Discipline:

Computer science

Sector:

Information and cultural industries; Manufacturing

University:

École de technologie supérieure

Program:

Accelerate

Current openings

Find the perfect opportunity to put your academic skills and knowledge into practice!

Find Projects