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

Motion tracking method based on composite deep neural network

A deep neural network and motion tracking technology, applied in the field of motion tracking of objects, can solve the problems of slow adjustment of weights, insufficient adjustment of the weights of the lower layers of the network, etc. The effect of tracking speed up

Inactive Publication Date: 2016-06-08
YANSHAN UNIV
View PDF2 Cites 13 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

When the number of network layers is deep, the network weight adjustment speed of the lower layer is very slow), and the weight value of the lower network layer is often not fully adjusted

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
  • Motion tracking method based on composite deep neural network
  • Motion tracking method based on composite deep neural network
  • Motion tracking method based on composite deep neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0029] Below in conjunction with the motion tracking process of video sequence " woman ", the inventive method is further described:

[0030] The present invention uses a motion tracking method based on a compound deep neural network, such as figure 1 As shown, its specific content includes the following steps:

[0031] Step 1 offline training:

[0032] Due to the need to track different objects, it is necessary to quickly adapt to the weights of different targets before using the observation model, so it is necessary to use a large number of samples for offline training of its weights; because the over-complete base vector consumes a lot of calculations during training and tracking Quantity, so the overcomplete basis vector is removed to simplify the structure of the network, and the neural network with decreasing number of nodes is used;

[0033] Step 2 Model initialization:

[0034] Add a logistic classifier to the top of the network obtained from offline training, so th...

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 motion tracking method based on a composite deep neutral network. The motion tracking method comprises the steps of performing offline training on a weight by means of a large number of samples; using a neural network of which the number of nodes reduce gradually; adding a logistic classifier at the top of the network which is obtained through offline training so that a target and a background can be distinguished, and setting the parameter of an adjusting network; adjusting an observation model in a tracking process; at the first frame, performing adaptive adjustment on the first frame, adjusting the network by means of the target and a background sample so that the network can identify the target; combining the observation model with a dynamic model in which a particle filtering algorithm is used; acquiring particles around the target of a previous frame in a new frame in particle filtering, transmitting the acquired particles to the observation model, determining the confidence of the particles by the observation model, and determining the particle with highest confidence as the target.

Description

technical field [0001] The invention relates to a motion tracking method of an object, in particular to a motion tracking method based on a compound deep neural network, which is a tracking method for detecting a target by using a deep neural network. Background technique [0002] Motion tracking is considered to be a very challenging task due to the influence of occlusion, light intensity changes, in-plane rotation, out-of-plane rotation, background clutter, etc., and is also an important part of computer vision. Motion tracking can be widely used in many fields, such as video surveillance, intelligent transportation, product detection, abnormal behavior detection. Although a large number of models have been proposed, several key issues are still not fully resolved. [0003] A motion tracking system generally consists of two models: an observation model and a dynamic model. The observation model is a model that describes the target object. Dynamic models are used to dete...

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 Applications(China)
IPC IPC(8): G06T7/20
CPCG06T2207/10016G06T2207/30196
Inventor 闻佳卢海涛赵纪炜
Owner YANSHAN UNIV
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