Method for searching visual lens based on contents

Abstract

The invention discloses a method of retrieval of a lens based on the content. Firstly, video database is splitted by the lens and the lens is taken as a basic structure unit and a retrieval unit; care areas are selected to represent the content of the video lens by a key frame strategy of the number of the key frame and a selective model of space-time attention; a lens A and a lens B are designed as significant areas and the matching of the two corresponding key frames is calculated; the matching of the lens space-time content is calculated. The representation mode of the lens content of the invention keeps the converted information with associated semanteme and context to be on a time line; then the video content is retrieved on the content representation, good retrieval effect is obtained and higher accuracy can be obtained.

Description

a technical field [0001] The invention belongs to the technical field of video retrieval, in particular to a content-based video retrieval method. Two background technology [0002] With the continuous development of computer technology, computer processing power has been significantly improved. The advanced integrated manufacturing process has continuously improved the hardware performance, and the innovation of network technology and communication technology has effectively promoted the development of information technology. The amount of media information processed by people's applications is increasing exponentially every day, and the development of information technology has brought massive video data. New forms of media information exchange and application such as video on demand, digital interactive TV, and video conferencing have been fully integrated into people's daily study, work, life, and entertainment. Digitized media information, especially digitized video a...

AI Technical Summary

Problems solved by technology

Digitized media information, especially digitized video and audio information has massive data, which causes great difficulties to the management, storage and transmission of information, and becomes the main bottleneck that hinders the effective acquisition and use of information by human beings.

According to the traditional text data analysis, management, and retrieval methods to manage video information, it will consume a lot of manpower and cannot make full use of the rich information contained in the video.

Therefore, traditional text information content description, analysis, management, retrieval and other related technologies are no longer suitable for processing large amounts of video data.

Images