Welcome to the Multimedia Laboratory

The Multimedia Laboratory at the University of Toronto, is part of the Communications Group at the Edward S. Rogers Sr. Department of Electrical and Computer Engineering. Our Laboratory has been at the forefront of the signal and image processing field. Specifically, research has been focused in the areas of biometric systems, secure and privacy enhancing multimedia solutions, nonlinear signal and image processing, multichannel image processing, morphological filters, neural networks and image and video coding.

Professor K. N. Plataniotis, Multimedia Lab Director

Project Highlights

eDREAM: Enhancing Driver inteRaction with digital mEdiA through cognitive Monitoring

In the last decade, digital media such as cell phone, FM radio, auxiliary music player, navigation system, and digital controls has had rapid growth in usage by drivers. Close at hand and useful, these technological advancements, however, have also imposed risks to drivers in the form of distraction. To better understand various aspects of driver distraction, the project will investigate what physiological features are affected by arousal, how arousal is defined in terms of physiological measures, and how these measures would be integrated into enhancing driving experience.
» Read more...

Automated Face Analysis by Feature Tracking and Expression Recognition

The face is an important information source for communication and interaction. This project aims to provide a robust facial feature tracking method based on active shape models and develop convolutional neural networks for a facial expression recognition task. The applications developed in this project achieve satisfactory performance.
» Read more...

Wireless Multi Input Multi Output (MIMO) Channel Simulation Package

The project aimed to provide a MATLAB software package for simulating wireless MIMO communication channels under practical assumptions. The final software provides an accurate simulation test bed which incorporates the effect of various pulse-shapes, fading models, noise structures, channel estimators, synchronization strategies, etc.
» Read more...

Learning for Biometric Signal Recognition

Biometrics is an important component in security-related applications such as access control, forensic investigation, and identity fraud protection. We focused on the very important problem of feature extraction through subspace learning in biometric systems.
» Read more...

Privacy Protected Surveillance Using Secure Visual Object Coding

The Secure Shape and Texture Set Partitioning in Hierarchical Trees (SecST-SPIHT) secure visual object coder allows individual, arbitrarily shaped objects to be efficiently coded (compressed) and encrypted. We use this in surveillance applications to protect the privacy of individuals appearing in the surveillance footage.
» Read more...

Multilinear Subspace Learning (MSL)

This project aims to provide an overview of resources concerned with theories and applications of multilinear subspace learning (MSL). The origin of MSL traces back to multi-way analysis in the 1960s and they have been studied extensively in face and gait recognition. With more connections revealed and analogies drawn between multilinear algorithms and their linear counterparts, MSL has become an exciting area to explore for applications involving large-scale multidimensional (tensorial) data as well as a challenging problem for machine learning researchers to tackle.
» Read more...