Our laboratory works in four main areas:
- Face and Image Analysis: We create feature descriptors for face and general images. Such descriptors are used to detect persons identities and their expressions. For the general images, we create enhancement algorithms, that ranges in a wide area, from brightening dark areas, to reducing the blockyness in video sequences.
- Expert Systems: We create MTES, that is an environment that allows teaching and learning image processing algorithms in a simple and reliable place. It is easy enough to introduce the field to new students, and powerful to do research on it. Moreover, we work on intelligent filters, which are algorithms that can adapt to the conditions of the image and fix it accordingly.
- Activity Recognition: We work on the detection and tracking of persons and objects. Our goal is to detect human activity from videos, and being able to track the persons performing such activities.
- Medical Image Processing: We work on orthodontics software. Our goal is to automatically segment the tooth from CT images, and build 3D models automatically. Such 3D visualizations can be used for medical purposes, showing the treatment desire state to a patient, and to produce prosthesis and orthodontics action plans.