With the vast development of Internet and media technologies, intelligent video event analysis is a gradually growing field of research in recent decades. Significant challenges include how to handle background clutter, occlusions, as well as interactions. Another difficulty is the lacking of widely accepted definition of event in the literature. Though the research achievement is still far from its promise, steady progress has been made in past years. This book collects a set of selected contributions in this area from international experts including leading academic researchers, and industrial practitioners. It presents the latest advances of intelligent video event analysis in both theoretical and application viewpoints.
Topics and features:
· - Addresses the concept of events by introducing a double view of understanding meaningful events in gesture based interaction
· - Investigates motion segmentation based on the subspace technique by incorporating the cues from the neighbourhood of intensity edges of images
· - Provides the state of the art techniques on human action description, and recognition based on 3D spatial temporal features
· - Presents efficient object localization and detection approaches in challenging scenes
· - Describes motion analysis techniques in various applications including sports videos, household environment, and surveillance videos
It provides researchers and practitioners a rich resource for future research directions and successful practice. It could also serve as a reference tool and handbook for researchers in a number of applications including visual surveillance, human-computer interaction, and video search and indexing, etc. Graduate students working on video analysis in various disciplines such as computer vision, pattern recognition, information security, artificial intelligence will also find it useful.