Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

An Action Recognition Method Based on Multi-instance Markov Model

A technology of Markov model and recognition method, which is applied in the direction of character and pattern recognition, instruments, computer parts, etc., to achieve the effect of reducing labeling

Inactive Publication Date: 2016-09-21
INST OF AUTOMATION CHINESE ACAD OF SCI
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to solve the problem of behavior recognition in complex scenes. To this end, the invention provides a behavior recognition method based on a multi-instance Markov model

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
  • An Action Recognition Method Based on Multi-instance Markov Model
  • An Action Recognition Method Based on Multi-instance Markov Model
  • An Action Recognition Method Based on Multi-instance Markov Model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0019] In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be described in further detail below in conjunction with specific embodiments and with reference to the accompanying drawings.

[0020] figure 1 It is a flow chart of the behavior recognition method based on the multi-instance Markov model proposed by the present invention, such as figure 1 As shown, the described behavior recognition method based on multi-instance Markov model comprises the following steps:

[0021] Step S1, establish a training set, the training set contains positive and negative sample videos of a certain action A; extract its local spatiotemporal interest points for each video in the training set; perform the first aggregation on the extracted spatiotemporal interest points Class, get these spatio-temporal interest points corresponding to the category of the first cluster center;

[0022] For example, clustering the extracted...

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 behavior recognition method based on a multi-instance Markov model. The method includes the following steps: extract local features for each video, and use a feature histogram of a local video block to represent a certain local motion of the behavior; obtain many local video blocks by random sampling, and these local video blocks will form multiple Markov chains, these Markov chains are expressed as continuous actions of some local motions in time; under the framework of multi-instance learning, the model selects the most discriminative Markov chain to represent the behavior; when testing, it is represented by In the same way, multiple Markov chains are formed to represent the video, and then the scores of these Markov chains are calculated. If it is greater than a certain threshold, it is this kind of behavior, otherwise it does not belong to this kind of behavior. The present invention uses a multi-instance Markov model to achieve the purpose of behavior recognition in complex scenes, and can reduce annotations on videos.

Description

technical field [0001] The invention belongs to the technical field of intelligent video monitoring, and in particular relates to a behavior recognition method based on a multi-instance Markov model. Background technique [0002] Behavior recognition has a wide range of applications in video intelligent surveillance. For example, in some specific occasions, airports, squares, streets, shops, etc., the early warning of dangerous behaviors of individuals and groups has important application value for public safety. At present, the mainstream behavior recognition is mainly through the construction of word bags based on local features. This method does not consider the distribution information and semantic information of behavioral feature points in time and space, and these local feature detectors detect many noisy feature points from the background. In order to solve these problems, Sadanand et al. proposed a new behavior representation method, which incorporates semantic in...

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): G06K9/62
Inventor 王春恒周文肖柏华张重
Owner INST OF AUTOMATION CHINESE ACAD OF SCI
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com