Human body target part automatic analytic method based on multiple images

Abstract

The invention discloses a human body target part automatic analytic method based on multiple images, and belongs to the field of computer vision and image processing. According to the method, on the basis that the multiple human body target images are given, by means of concurrence among targets, human body target parts in the images are automatically analyzed. By means of the method, a human body target classifying module and a human body target analyzing module are executed in an iterated mode until the stopping condition is reached. The human body target classifying module achieves automatic classification of the multiple human body target images. In a classifying process, appearance similarity and gesture similarity among the human body targets are considered comprehensively; the human body target analyzing module achieves automatic analysis of the human body target parts in each classification and alignment of analyzed results of different classifications. According to the method, independent analysis of each human body target image is converted into simultaneous analysis of the multiple human body target images, and the limitation caused by a traditional analysis mode that learning is conducted and then inference is conducted is overcome.

Description

[0001] technical field [0002] The invention belongs to the field of computer vision and image processing. Specifically, the invention relates to a method for automatically realizing analysis of human body target parts in multiple images, and is a basic technology for applications such as human body target modeling. Background technique [0003] Object parsing in images is an important branch of computer vision and image understanding, and is often used to realize object recognition, segmentation and pose estimation in images. Traditional image object parsing techniques require two key steps. A critical step is the offline modeling process of target part annotation. Another critical step is the online inference process of target component parsing. In the offline process, the definition of the target part and the labeling of the target part need to be performed. The definition of target parts is to determine which basic elements a target consists of. Almost all previous wo...

AI Technical Summary

Problems solved by technology

[0006] The problem solved by the present invention is: overcoming the problem that the definition of human body target parts in the traditional human body target part analysis technology is difficult to correspond to the actual area in the image and the high consumption and uncertainty caused by manual labeling, the present invention proposes An automatic analysis method for human body target parts based on multiple images, which uses recurring patterns to automatically define and analyze human body target parts, comprehensively considers the shape information and posture information of human body targets, and analyzes the target from the perspective of part combination mode Quantification has the advantage of distinguishing human targets with different postures, achieves the effect of discovering significant human target parts, and realizes the function of analyzing multiple related targets at the same time

Images