天美传媒

Sergio Alejandro Velastin Carroza

Sergio Alejandro Velastin Carroza
Kingston University UK

Prof. Sergio A Velastin has worked for a number of years on visual monitoring of people and vehicles and the detection of incidents related to security and safety.

He has had national and EU funding (CROMATICA, ADVISOR, PRISMATICA (Technical Co-oordinator), CARETAKER, ADDPRIV and ProtectRail). He has also led industrially-funded research for the railways, transport authorities and a major global manufacturer.

In 2003 he conceived and still organises and chairs the international conference on Imaging for Crime Detection and Prevention (ICDP). He has authored more than 100 publications on visual surveillance and until recently he was the director of the Digital Imaging Research Centre in Kingston University London.

He is currently a research professor at the University of Santiago (Chile). He is a Senior Member of the IEEE and a Fellow of the IET and regularly gives talks on video analytics to scientific and commercial audiences.

RESEARCH STAY AT UC3M: DEPARTMENT OF COMPUTER SCIENCE

Project: The general goal for the project is to investigate advanced machine vision algorithms and systems in relevant applications (surveillance systems, ambient intelligence, intelligent vehicles), integrated into an extended framework so that new disciplines in computational intelligence such as context exploitation, knowledge modeling, distributed reasoning, machine learning, etc. can be explored and analyzed. Work will take place in collaboration with a number of groups at Carlos III (GIAA, EVANNAI and CAOS at Department of Informatics, SLI at Department of Automatics and Systems Engineering, and Department of State Public Law) and will include machine algorithms in the domain of robust people detection and tracking, formalization of the use of contextual information to enhance system performance e.g. for assisted living applications, development of architectures for sensor coordination and management in distributed vision, demonstration of real-time performance and integration of a variety of sensors.

Stay period: ENE 2014 - JUL 2014