Enhancement coding method for visual mapping target value

A technology to enhance coding and target value, applied in the direction of instrument, character and pattern recognition, acquisition/recognition of eyes, etc., can solve the problems of poor visual mapping estimation effect, uneven distribution, sparse samples, etc., to improve the sample recognition rate. Effect

Active Publication Date: 2017-03-15
济南世纪优势信息科技有限公司
View PDF7 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This invention patent solves the problem that the existing methods of visual mapping have poor estimation results when the samples are sparse and unevenly distributed.

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
  • Enhancement coding method for visual mapping target value
  • Enhancement coding method for visual mapping target value
  • Enhancement coding method for visual mapping target value

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0039] Implementation language: Matlab, C / C++

[0040] Hardware platform: Intel core2 E7400+4G DDR RAM

[0041] Software platform: Matlab2012a, VisualStdio2010

[0042]According to the method of the present invention, first clearly the visual mapping problem to be solved, and collect relevant images (head image, body image and facial image, etc.) and calibrate the target value (head posture angle, body posture angle and age). According to the patent of the present invention, first use Matlab or C language to write a program to learn the mapping model from the image to the enhanced code, and the random forest model from the enhanced code to the target value; then visually map the input image to be estimated to estimate the target value. The method of the invention can be used for visual mapping problems in various computer visions, and obviously improves the performance of the direct mapping method (from input features to target values).

[0043] The present invention further...

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, which belongs to the technical field of the computer vision, especially to the visual mapping technology, provides an enhancement coding method for a visual mapping target value. An image is collected and feature extraction is carried out, a corresponding target value is recorded; enhancement coding is carried out on the target value, wherein each coded bit is a 0 / 1 binary variable; a mapping relation between an original input image feature and binary coding is established; all input images are mapped to the binary coding based on the mapping relation; and then a mapping relation between binary coding and the target value is established by using a random forest method. For a new test picture, image feature is extracted; binary coding is estimated by using the learned model; and the binary coding is regressed to the target value. Under the circumstance that the samples are sparse and are not distributed uniformly, the sample identification rate as well as the identification precision can be improved.

Description

technical field [0001] The invention belongs to the technical field of computer vision, relates to visual mapping technology, and is mainly applied to visual estimation problems such as attitude estimation, line of sight tracking and age estimation. Background technique [0002] In computer vision, visual mapping refers to the process of learning a mapping function between input image features and output variables, so that when a new image is input, the target output value corresponding to the input image is estimated. Specifically, visual mapping includes: human body pose estimation, head pose estimation, line of sight estimation, and object tracking. See references for details: O.Williams, A.Blake, and R.Cipolla, Sparse and Semi-Supervised Visual Mapping with the S3GP, in IEEEConference Computer on Computer Vision and Pattern Recognition, pp.230-237, 2006. [0003] As an important branch of computer vision, visual mapping has changed the situation where humans estimate ta...

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
IPC IPC(8): G06K9/00G06K9/46G06K9/62
CPCG06V40/178G06V40/16G06V40/20G06V40/19G06V10/446G06V10/40G06V10/50G06V10/462G06F18/231G06F18/214
Inventor 潘力立
Owner 济南世纪优势信息科技有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap