Flexible, semi-automated video forensics system for the efficient analysis of mass video data

Video material collected and analysed by law enforcement agencies (LEA) has become a critical component in legal investigations following major criminal acts and terrorist attacks. The recent events in London, Nice, Brussels, Paris, Barcelona and Cambrils again prove the importance of these video recordings. At the same time, the amount of video data available is continuously increasing with the rapid deployment of video equipment, surveillance cameras in public and private areas, body-worn cameras of police forces and smartphones or digital cameras used by bystanders.

In spite of this growth, the whole video investigation work is still mostly carried out manually by the LEA officers. These current practices are too resource intensive to handle the huge and steadily increasing volume of videos that need to be analysed. Consequently, post-event extraction of vital first clues from videos meet unreasonable delays. Time is a precious factor for LEAs thus there is an urgent need for efficient tools assisting them in their daily video investigation tasks.

For addressing this need, the Austrian research centre AIT Austrian Institute of Technology is working within national and European research projects on a flexible, semi-automated video forensics system for the analysis of mass video data after terrorist attacks. This system will improve the efficiency and quality of work for investigators.

Furthermore, AIT is contributing their combined expertise in video and audio analysis, as well as in the use of artificial intelligence models in order to develop an intelligent and special forensics platform. This will allow very large and unstructured streams of raw data from videos, images, audio recordings and documents to be intelligently filtered, sorted and automatically annotated , so that the investigating officers can very efficiently create a digital case file.

• efficient, open and flexible tool to assist LEAs in their daily video investigation tasks
• video and audio analysis, based on artificial intelligence-models to allow very large and unstructured streams of raw data from videos, images, audio recordings and documents to be intelligently filtered, sorted and automatically annotated
• analytics robust to camera characteristics, image quality, viewing angles, moving cameras, etc.