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

A Behavior Recognition Method Based on 3D Convolutional Neural Network

A convolutional neural network and recognition method technology, applied in the fields of pattern recognition and video image processing, machine learning, and feature matching, can solve the problem of lack of classification ability for short-term simple actions, achieve high accuracy and avoid overfitting , the effect of a small amount of calculation

Inactive Publication Date: 2017-11-17
UNIV OF ELECTRONICS SCI & TECH OF CHINA
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Another method is to use storyline to describe the occasional relationship between behaviors, and OR graphs (AND-OR graphs) are used as a mechanism to represent the storyline model, which lacks the ability to classify short-term simple actions

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 Behavior Recognition Method Based on 3D Convolutional Neural Network
  • A Behavior Recognition Method Based on 3D Convolutional Neural Network
  • A Behavior Recognition Method Based on 3D Convolutional Neural Network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0030] The BP algorithm is used for training, but the network structure of CNN itself is very different from the traditional neural network, so the BP algorithm used by CNN is also different from the traditional BP algorithm. Since CNN is mainly composed of convolutional layers and downsampling layers alternately, their respective formulas for calculating the reverse error delta propagation are different.

[0031] Using the square error cost function, the calculation formula of the output layer δ is:

[0032]

[0033] Among them, y is the actual output vector of the network, t is the expected label vector, there are n components, and the f function is a sigmoid function. is the Schur product, that is, the multiplication of the corresponding elements of the two vectors, u is the weighted sum of the output of the upper node, and the calculation formula is as follows:

[0034] u l =W l x l-1 +b l

[0035] The output x of layer l-1 is multiplied by the weight W of layer ...

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 3D convolutional neural network, and relates to the fields of machine learning, feature matching, pattern recognition and video image processing. The method is divided into two stages: first, offline training, by inputting sample videos of various behaviors, and calculating different outputs, each output corresponds to a type of behavior, and then correcting the calculation process according to the error between the output vector and the label vector The medium parameter reduces the error of each output data. After the error meets the requirements, add labels to each output according to the behavior name of the corresponding sample video; secondly, perform online recognition, input the video that needs behavior recognition, and use the same method as the training stage to calculate Then output, and then match the output with the tagged sample vector, and regard the name of the sample tag that best matches it as the behavior name of the input video, so it has low complexity, small amount of calculation, high real-time performance, and high accuracy. high effect.

Description

technical field [0001] The invention relates to the field of computer vision, in particular to methods for machine learning, feature matching, pattern recognition and video image processing. Background technique [0002] Behavior recognition by computer is to understand and describe human behavior from video or image sequences containing people, which belongs to the category of image analysis and understanding. The ability to automatically detect people and understand their behavior is a core function of an intelligent video system. In recent years, there has been an increasing interest in human behavior recognition due to the needs of society, including industrial security, switching interfaces, games, etc. The research content of human behavior recognition is very rich, mainly involving pattern recognition and machine learning, image processing, artificial intelligence and other subject knowledge. Three existing mainstream technical solutions for behavior recognition are...

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): G06K9/62G06N3/02
CPCG06V40/23G06V20/46G06V10/28G06F18/2411
Inventor 郝宗波桑楠吴杰余冬
Owner UNIV OF ELECTRONICS SCI & TECH OF CHINA