6 June 2013 Recognition and localization of relevant human behavior in videos
Author Affiliations +
Ground surveillance is normally performed by human assets, since it requires visual intelligence. However, especially for military operations, this can be dangerous and is very resource intensive. Therefore, unmanned autonomous visualintelligence systems are desired. In this paper, we present an improved system that can recognize actions of a human and interactions between multiple humans. Central to the new system is our agent-based architecture. The system is trained on thousands of videos and evaluated on realistic persistent surveillance data in the DARPA Mind’s Eye program, with hours of videos of challenging scenes. The results show that our system is able to track the people, detect and localize events, and discriminate between different behaviors, and it performs 3.4 times better than our previous system.
© (2013) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Henri Bouma, Henri Bouma, Gertjan Burghouts, Gertjan Burghouts, Leo de Penning, Leo de Penning, Patrick Hanckmann, Patrick Hanckmann, Johan-Martijn ten Hove, Johan-Martijn ten Hove, Sanne Korzec, Sanne Korzec, Maarten Kruithof, Maarten Kruithof, Sander Landsmeer, Sander Landsmeer, Coen van Leeuwen, Coen van Leeuwen, Sebastiaan van den Broek, Sebastiaan van den Broek, Arvid Halma, Arvid Halma, Richard den Hollander, Richard den Hollander, Klamer Schutte, Klamer Schutte, "Recognition and localization of relevant human behavior in videos", Proc. SPIE 8711, Sensors, and Command, Control, Communications, and Intelligence (C3I) Technologies for Homeland Security and Homeland Defense XII, 87110B (6 June 2013); doi: 10.1117/12.2015877; https://doi.org/10.1117/12.2015877

Back to Top