Computational Pattern Analysis

Click on the links below for more details.

Telepresence

Remote collaborative working environments aim at reducing time and space constraints by exploiting existing computer and network infrastructures. Such environments become popular in business, education and research institutions. Several integrated environments for supporting collaborative work exist, but there are very few solutions which consider the hardware as well as the software sharing dimension of collaborative workspaces. Research conducted at Synchromedia ultimately resulted in highly integrated working platform supporting remote collaborative work.

Intelligent Interfaces

The bottleneck in human-machine interaction is exactly to make the human user and a computer system interact. The system must adapt to the user, and the user to the system. Usually, user interfaces shift the entire burden of adaptation to the user. Typically, intelligent interfaces will attempt to adapt to users and model knowledge about them.

Ongoing research at Synchromedia address the development of intelligent interfaces integrating new perceptual and interacting devices in order to enhance user experiences and improve collaborative working environments. We have been particularly active with mobile touch and pen-based devices (interactive white boards, Tablet PCs, smartphones, digital pens, etc.), and the understanding of handwriting contents over collaborative sessions.

Biomedical Image Processing

Information processing is now one of the main frontiers in medical imaging. Modern medical imaging modalities revolutionize many aspect of the practice of medicine. Software tools have potentials in assisting practitioners in data interpretation, therefor lowering human-induced bias, reducing cost and even enabling new applications.

Synchromedia is active in the development of medical image processing methods in the following fields of applications: 3D/2D registration, segmentation of 3D vascular structure in CTA volumes, segmentation of brain tissues in MRI volumes, and de-noising and enhancement filters for application in orthopedics.

Document Processing and Understanding

Processing of huge volumes of unprocessed, handwritten and historical documents is a critical challenge in front of many heritage and cultural institutes and organizations. Our main objective in the field of document image processing and understanding is development and implementation of novel models and techniques which may help in generating, enhancing, presenting and understanding of handwritten document images. Direct involvement of scholars and researchers from various institutes and universities, such as McGill University, allowed us to choose goal-oriented directions for our research and development. Currently, our focus is on providing a complete understanding system which consists of imaging, pre-processing/enhancement, word-spotting, transliteration and data mining units in an user-friendly collaborative and virtual environment.

Databases and Contests

In order to train the learning machines of the binary problems associated to OSR, two databases of different sizes have been created in collaboration with Prof. Robert Wisnovsky (Institute of Islamic Studies, McGill University).

Classification and Machine Learning

Machine learning is a field of artificial intelligence, which use empirical data to extract characteristic of interest of their unknown underlying probability distribution. Knowledge about a given problem is learnt from examples (training data). This knowledge is implicit, unlike explicit knowledge that can be given by formal definitions or rules. A machine learning algorithm is evaluated on its generalization capability, which is its ability to apply successfully the learnt knowledge to unseen data. Different kinds of machine learning algorithms exist, depending on the goal. Supervised algorithms such as support vector machine (SVM) are used for classification tasks while unsupervised algorithms such as manifold learning algorithms are used to extract the relation between the examples.

Experimental Results

Experimental results of the published papers, including the full size images, could be found in this section. Usually, there is a permanent link in each paper that points directly to its experimental results.You can also browse in this section.
ericssonlogo
inocybelogo
canalogo
cienalogo
Civimetrix Telecom logo
mitacslogo
risq logo
nserclogo
promptlogo
ecolepolytechniquelogo
University of Torontologo
frqntlogo
uqlogo
MDEIE logo
cfilogo
ciraiglogo