Method and system for motion vector-based video monitoring and event categorization

作者: Jason N. Laska , Greg Duffy , Gregory R. Nelson

DOI:

关键词: CategorizationVideo monitoringMotion (physics)Object (computer science)Motion vectorComputer scienceEvent (relativity)Artificial intelligenceTrack (disk drive)Computer visionEvent recognition

摘要: A computer system processes a video stream to detect start of first motion event candidate in the stream, and response detecting initiates recognition processing on segment associated with candidate. Initiating further includes: determining track object identified segment; generating representative vector for based object; sending an categorizer, where categorizer assigns respective category

参考文章(291)
Tobias Delbrück, Frame-free dynamic digital vision Delbruck, T (2008). Frame-free dynamic digital vision. In: International Symposium on Secure-Life Electronics, Tokyo, Japan, 6 March 2008 - 7 March 2008, 21-26.. ,vol. 1, pp. 21- 26 ,(2008) , 10.5167/UZH-17620
Jeffrey E. Skinner, Xin Feng, Aaron Michael Stewart, Special gestures for camera control and image processing operations ,(2013)
Kenneth M. Carson, Giovanni Agnoli, Eric J. Graves, Shaun M. Poole, Media-Editing Application with Video Segmentation and Caching Capabilities ,(2011)
Wonpil Yu, Jae-Yeong Lee, Michael Sahngwon Ryoo, Method for human activity prediction from streaming videos ,(2012)
Russell P Bobbitt, Rogerio S Feris, Yu Cheng, Yun Zhai, Determination of train presence and motion state in railway environments ,(2015)