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

A Face Portrait Recognition Method Based on Face Attributes

A face portrait and recognition method technology, applied in the field of image processing, can solve the problems of low recognition accuracy, insufficient information, and failure to use face attribute information coding features, etc., to achieve high recognition accuracy and strong resolubility

Active Publication Date: 2022-02-22
XIDIAN UNIV
View PDF14 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] However, the above two methods do not use face attribute information when using features to encode, and the encoding feature information used is insufficient, resulting in low recognition accuracy

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 Face Portrait Recognition Method Based on Face Attributes
  • A Face Portrait Recognition Method Based on Face Attributes
  • A Face Portrait Recognition Method Based on Face Attributes

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0035] See figure 1 , figure 1 It is a schematic flowchart of a face portrait recognition method based on face attributes provided by an embodiment of the present invention. A method of face portrait recognition based on face attributes provided by an embodiment of the present invention includes:

[0036] Obtain training sample set and test sample set;

[0037] Utilizing the training sample set to train a deep human face attribute representation model;

[0038] Using the deep face attribute representation model to obtain the face attribute representation features of the test sample set;

[0039] The similarity calculation is performed by using the face attribute representation features to identify the face portraits in the test sample set.

[0040] Specifically, the training sample set is used to train the deep face attribute representation model.

[0041] The test sample set is used to test the trained deep face attribute representation model, so as to determine the reco...

Embodiment 2

[0046] On the basis of the above-mentioned embodiments, the embodiment of the present invention specifically introduces the face portrait recognition method provided by the embodiment of the present invention. The recognition method specifically includes:

[0047] Step 1. Obtain a training sample set and a test sample set;

[0048] Specifically, the training sample set includes N groups of first sample groups, wherein each group of first sample groups includes a first face portrait, a first photo with the same identity as the first face portrait, and a For a second photo with a different identity from the first face portrait, the test sample set includes M groups of second sample groups, wherein each group of second sample groups includes a second face portrait and a photo with the same identity as the second face portrait. In the same third photo, M and N are natural numbers greater than 0.

[0049]Among them, having the same identity means the same person, and having differ...

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 present invention relates to a human face portrait recognition method based on human face attributes, comprising: obtaining a training sample set and a test sample set; using the training sample set to train a deep human face attribute representation model; using the trained deep human face The attribute representation model acquires the face attribute representation features of the test sample set; uses the face attribute representation features to perform similarity calculation to identify the second face portrait in the test sample set. The recognition method of the present invention uses a deep face attribute representation model to extract features from face portraits and photos, and overcomes the problem that the encoding feature information does not consider face attribute information when encoding face portraits and photos in the prior art, making the present invention The invention can obtain the attribute information of the human face, so that the recognition accuracy is higher, and the discrimination of the human face portrait is stronger.

Description

technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to a face portrait recognition method based on face attributes. Background technique [0002] In the process of criminal investigation and case solving, it is generally difficult to obtain photos of criminal suspects. Generally, the face portraits of criminal suspects are drawn through the descriptions of witnesses or victims for subsequent face retrieval and recognition. Due to the great difference between face portraits and ordinary photos, it is difficult to obtain satisfactory recognition results by directly using traditional face recognition methods. Face portrait-photo recognition technology improves the recognition rate by reducing the difference between two different images, and has received extensive attention in the field of image processing. [0003] A.Alex and others in the literature "A.Alex, V.Asari, and A.Mathew. Local difference of Gaussian bina...

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): G06V40/16G06V10/774G06V10/82G06K9/62G06N3/04G06F16/583
CPCG06V40/168G06N3/045G06F18/214
Inventor 高新波王楠楠刘德成李洁彭春蕾朱明瑞曹兵马卓奇辛经纬郝毅
Owner XIDIAN 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