Unlock instant, AI-driven research and patent intelligence for your innovation.

A learning method and system for human body appearance model for human body reproduction detection

An appearance model and human body reproduction technology, applied in the field of pattern recognition and computer vision, can solve problems such as weak ability, achieve the effect of improving accuracy rate, strong robustness, and increasing the number of training samples

Active Publication Date: 2016-09-07
INST OF COMPUTING TECH CHINESE ACAD OF SCI +1
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Traditional methods usually find some background parts as counterexample samples, which makes the learned appearance model weaker in distinguishing different human bodies

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A learning method and system for human body appearance model for human body reproduction detection
  • A learning method and system for human body appearance model for human body reproduction detection
  • A learning method and system for human body appearance model for human body reproduction detection

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0050] Specific embodiments of the present invention are given below, and the present invention is described in detail in conjunction with the accompanying drawings.

[0051] Such as figure 2 As shown, a learning method for a human appearance model for human body recurrence detection includes the following steps:

[0052] Step 1, extract the appearance features of the human body in the detection window of the video frame, and establish the appearance descriptor of the human body; the descriptor means the feature vector, usually in the form of a real number vector,

[0053] Step 2: Extract positive and negative human appearance descriptors based on spatio-temporal constraints for training the human appearance model.

[0054] The described learning method of the human body appearance model that is used for human body recurrence detection also includes in step 2:

[0055] Updating of the human appearance model of the human body online at the same time and / or updating of the hu...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a human body appearance model learning method and system for human body reappearance detection. The human body appearance model learning method includes the following steps: S1, extracting human body appearance characteristics in a detection window of a video frame, establishing a human body appearance descriptor and S2, extracting a positive example and a negative example of the human body appearance descriptor on the basis of space-time constraint information, wherein the positive example and the negative example are used for training the human body appearance model.

Description

technical field [0001] The invention belongs to the fields of pattern recognition and computer vision, in particular to a learning system and method of a human body appearance model for human body reappearance detection. Background technique [0002] The so-called human body reappearance detection is to find the image frame of a person under a camera at a certain moment (with the person's human body detection window), and find all the image frames that the person has appeared under the current camera or other cameras at other times / sequence. Human body reappearance detection can be accomplished by extracting the appearance model of the currently observed human body, and looking for a person who can well match the appearance model from image frames captured by the same or different cameras at other times in a model comparison manner. Since human body reappearance detection systems are usually affected by changes in various factors such as human body posture, viewing angle, ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/00G06K9/62
Inventor 王亮李浩常虹曾炜山世光陈熙霖
Owner INST OF COMPUTING TECH CHINESE ACAD OF SCI