Method for automatically detecting obvious object sequence in video based on learning

Abstract

The invention discloses an automatic inspection method of a significant object sequence based on studying videos. In the method of the invention, static significant features are firstly calculated, and then dynamic significant features are calculated and self-adaptively combined with the static significant features to form a significant feature restriction; the space continuity of each image of frame is calculated; the time continuity of significant objects in neighboring images is calculated. The similarity between all possible significant objects is calculated by the method; a significant object sequence obtained through the former calculation is utilized to calculate the overall subject model and calculate corresponding energy contribution; the overall optimum solution is solved by dynamic planning so as to obtain the overall optimum significant object sequence; the iteration is continued for solving if a convergence condition is not satisfied, otherwise a rectangle box sequence is outputted as the optimum significant object sequence. The method of the invention can effectively settle the choosing of the static and dynamic significant features, the optimum integration of various restraint conditions and the high effective calculation of target sequence inspections.

Description

technical field [0001] The invention belongs to visual attention analysis and automatic detection of salient object sequences in videos, in particular to an automatic detection method for a single salient object sequence. Background technique [0002] With the development of the Internet and digital camera technology, more and more videos can be obtained on the Internet, which makes video retrieval, video arrangement and summary very important. Content analysis in videos has always been an important part of video retrieval and summarization. Traditional video attention analysis identifies the importance of different parts of the video by assigning different weights to pixels in the spatio-temporal sequence, and can identify possible visual focus points, but there is no concept of "object" and thus cannot detect objects. The concept of objects in reality refers to things we often see, such as human faces, cars, pedestrians, tables, cats, dogs, etc., and salient objects refer...

AI Technical Summary

Problems solved by technology

[0005] The method of video attention analysis focuses on counting directly from features such as the surface of the image or the sports field. Although some attention points can be identified, it cannot solve the overall concept of "object". In addition, the information that does not integrate the whole world often brings calculation error

The biggest problem with the method of automatic object discovery in the video is: static and dynamic salient feature selection and effective integration, as well as the efficiency and convergence of the algorithm

Images