Study of Object Segmentation Motion in Dynamic Lookup Scene Video

• Main Authors: Wang Weihong, Wang Xiaojun
• Format: Article
• Language: English
• Published: IFSA Publishing, S.L. 2013-10-01
• Series: Sensors & Transducers
• Subjects: Image processing; Data structure; Algorithm; Topography; Diving.
• Online Access: http://www.sensorsportal.com/HTML/DIGEST/october_2013/P_1427.pdf

With the rapid development of computer technology, motion analysis of video has become a hot research topic in the field of computer vision based on. As the core technology and the underlying problems of visual analysis of moving objects, moving objects in video from a dynamic scene segmentation is very important, it is the subsequent high-level processing, such as moving object tracking, object classification and activity recognition, event detection based high level video processing and application of understanding. The paper introduces the segmentation technology to monitor the video in multimedia, video object, application and research situation of sports analysis and other fields, and discussed the significance and application of the high speed development of the information age, the development situation of domestic and foreign video segmentation techniques are reviewed, including spatial intra-frame segmentation, global motion estimation, a few hand joint space-time segmentation discussed video segmentation method, and according to the characteristics of diving video sequences, to study how to extract moving objects from a dynamic scene problem. On this basis, the video object extraction method used in diving and gymnastics videos.