Resource-Conscious Frameworks for Video Summarization and Prioritization
In recent years, there has been a tremendous increase in video capturing devices, which led to large personal and corporate digital video archives. This huge volume of video data became a source of inspiration for the development of vast numbers of applications such as visual surveillance, multimedia recommender systems, and context-aware advertising. The heterogeneity of video data, higher storage, processing cost, and communication requirements demand for a system that can efficiently manage and store huge amount of video data, while providing user-friendly access to stored data at the same time. To address this problem, video summarization can be considered as a pre-processing step which refers to the extraction of keyframes, identifying most important and pertinent content. For instance, gastroenterologist uses wireless capsule endoscopy video technology to diagnose his patients. However, during capsule endoscopy process, video data are produced in huge amounts, but only a limited amount of data is actually useful for diagnosis. Analysis of an endoscopy video is a long duration and tedious task, which limits the number of patient studies that can be performed daily. Consequently, degrades the efficiency of medical staff and effective utilization of facilities. In this context, summarization can play a vital role, estimating the semantically relevant video frames from the perspective of gastroenterologists. Similarly, in remote visual surveillance tasks, the sharing and analysis of surveillance videos becomes a challenging task due the constraints such as limited memory, energy, and communication capability. Video summarization service can select visual content with suspicious activities, sending only high priority frames to the security analysts instead of sending all the data. In various applications, video summarization can be conducted from the perspective of information prioritization, ranking chosen keyframes relative to their ability to describe the content of the video. A good video summary improves the effectiveness and efficiency of video archiving, cataloging, indexing, as well as increasing the usability of stored videos.
In this talk, video summarization in general and specifically in the context of prioritization (VSP) will be discussed. Several case-studies from different domains such as medical imaging, visual surveillance, disaster management, and entertainment industry will be discussed. Some research directions will be suggested as future works to further enhance the performance of the VSP techniques and investigate their ability to help experts organizing information of their particular interests in various computing domains.
Irfan Mehmood
Lecturer Applied Artificial Intelligence
Faculty of Engineering & Informatics, School of Media, Design and Technology
University of Bradford, Richmond Road,
Bradford BD7 1DP, UK.
Office: D 2.16, Horton Building