Method for extracting video sequence key frame in three-dimensional reconstruction

Abstract

The invention discloses a method for extracting a video sequence key frame in three-dimensional reconstruction. The method is used for extracting the key frame from a video sequence used for conducting three-dimensional reconstruction. According to the method, starting from a first frame of the video sequence, a frame of video image which is clear and comprises the content for three-dimensional reconstruction is selected to be used as a first key frame which is used as a standard frame, the comprehensive feature similarity of the standard frame and current frames of vide images is compared one by one, the video image with small redundant information is extracted to be used as the key frame according to the comparison result, and the newly extracted key frame is used as the standard frame to continuously conduct comparing until the last frame in the video sequence is compared. According to the method for extracting the video sequence key frame in three-dimensional reconstruction, the key frame can be effectively extracted, a large amount of redundant information can be removed, the operation time for three-dimensional reconstruction is shortened, and the precision of the three-dimensional reconstruction is improved.

Description

technical field [0001] The invention belongs to the technical field of three-dimensional reconstruction, in particular to a method for extracting key frames of a video sequence in three-dimensional reconstruction. Background technique [0002] With the practical application of computer vision technology in various fields, various information based on 3D has attracted more and more attention, such as digitally simulated cities, medical treatment using digital information, computer virtual animation (virtual reality), high-precision Research in cutting-edge science, creation in the field of high art, etc. From single-view viewing to multi-view interactive browsing, the enjoyment of high-fidelity three-dimensional stereo vision has gradually become the daily needs of people's lives, and the way people perceive information has also undergone earth-shaking changes. [0003] It is an important task to make full use of existing information to realize people's needs by organizing i...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to solve the problem that in the existing three-dimensional reconstruction process, because the video sequence has a large amount of redundant information, a lot of information is useless or even has the result of fuzzy matching accuracy, and proposes a method for extracting video sequences in three-dimensional reconstruction. The key frame method is based on deleting a large amount of redundant information and improving the running time and accuracy of 3D reconstruction

Images