| KEYWORD |
Target tracking in video analysis
Reference persons FABRIZIO LAMBERTI
Research Groups GR-09 - GRAphics and INtelligent Systems - GRAINS
Description Target tracking represent an image processing technique used to identify the same object or objects of interest throughout a video sequence. A variety of acquisition systems can be used, including surveillance systems (both single or multi-camera), high resolution videos, onboard sensors (for drones or cars), and so forth. Infrared cameras may also be used when particular environmental situations occurs, e.g. in case of fog and night vision. Uses cases of interest for the GRAINS group include, among others, surveillance, robotic vision, video interpretation and analysis, and robotic gaming. Multiple thesis are available in which the student will be asked to analyze, design and implement state-of-the-art systems for target tracking. The student(s) will be asked to analyze one or more applications of interest in order to (a) improve the robustness and/or speed of the system, or (b) integrate information coming from different sensors. Skills possessed or to acquire: programming skills (Python, Keras/Tensorflow or other deep learning framework), computer vision, machine learning, data fusion.
See also http://grains.polito.it/work.php
Deadline 18/05/2019
PROPONI LA TUA CANDIDATURA