Method and system for figure detection, body part positioning, age estimation and gender identification in picture of network

A technology for gender identification and body parts, which is applied in the field of person detection in network pictures, body part positioning, age estimation, gender identification and system fields, which can solve the problem that there is no more research on refined positioning.

Inactive Publication Date: 2013-09-11
北京明日时尚信息技术有限公司
View PDF3 Cites 21 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Faces or pedestrians can be detected through face detection technology and pedestrian detection technology, but there is no more research on the fine positioning of each part

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
  • Method and system for figure detection, body part positioning, age estimation and gender identification in picture of network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0005] The purpose of the present invention is to provide a method and system for person detection, body part location, age estimation and gender identification in network pictures, including: 1. training face classifiers on face samples; 2. training pedestrians on pedestrian samples Classifier; 3. Train AAM model data on face samples; 4. Train age and gender recognition classifiers on face samples of different genders and ages; 5. Load their respective classifiers to realize person detection in network pictures , body part positioning, age estimation and gender identification; the specific steps are as follows:

[0006] 1. Face classifier training process

[0007] Collect face positive and negative samples, do histogram equalization and normalize to 24×24 size;

[0008] Extract Haar-Like features from face samples;

[0009] Pass the face features into Adaboost training to get a face classifier;

[0010] 2. Pedestrian classifier training process

[0011] Collect positive a...

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 provides a method and system for figure detection, body part positioning, age estimation and gender identification in a picture of a network. The method includes the steps that by means of a face detection technique based on Haar-Like characteristics and Adaboost trainings and a pedestrian detection technique based on HOG characteristics and SVM trainings, a face area and an overall body area are primarily detected, and then a further fusion is carried out so that a face can correspond to a body if the body exists; active appearance models (AAM) are used in the face part so as to position the positions of the five sense organs of the face, and the position of the upper half body, the position of the lower half body, the positions of the left hand and the right hand, and the position of feet are confirmed on a body part according to a human body geometric model; eventually, identification of gender and age is carried out on the face part according to GIST characteristics and an SVM. Therefore, whether a figure exists in the picture of the network and various biological characteristics of the figure can be obtained through calculation.

Description

technical field [0001] The invention relates to a method and system for person detection, body part positioning, age estimation and gender recognition in network pictures, which belongs to the field of image recognition, and specifically performs face detection and upright body detection in static pictures, and then Position the hair, eyes, nose, cheeks, and mouth on the face, and estimate the age and gender of the face; position the upper body, lower body, feet, and left and right hands on the body. Background technique [0002] The pictures on the Internet are mainly people. Whether it is a news picture or a product picture, most of the pictures are related to people, that is to say, there are people in most of the pictures. Faces or pedestrians can be detected by face detection technology and pedestrian detection technology, but there is no more research on the fine positioning of each part. Based on the face detection technology and pedestrian detection technology, the ...

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): G06K9/00G06K9/62
Inventor 胡楠邹国平李洋
Owner 北京明日时尚信息技术有限公司
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