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

Method and device for identifying character trajectory in video

A video track and track technology, applied in the field of face recognition, can solve problems such as increased cost, unreasonable, and increased calculation, and achieve cost-saving and wide-ranging effects

Active Publication Date: 2017-09-01
SHENZHEN IPANEL TECH LTD
View PDF4 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

It can be seen that if the existing face recognition technology is used to study the trajectory of someone in the video, it is necessary to establish a stock of known faces and training samples, and the need for stock of known faces greatly reduces the performance of face recognition technology. The scope of application requires a large number of training samples. Although it can improve the recognition rate, it increases the cost, and the comparison with the known faces in the inventory also greatly increases the amount of calculation.
Obviously, it is unreasonable to rely on existing face recognition technology for this purpose

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 device for identifying character trajectory in video
  • Method and device for identifying character trajectory in video
  • Method and device for identifying character trajectory in video

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0050] see figure 1 It is a schematic flow chart of a method for identifying the trajectory of a person in a video provided by Embodiment 1 of the present invention. The method includes the following steps:

[0051] S11. Convert all video segments containing human face images in the video to be identified into a plurality of color pictures containing human face images;

[0052] Specifically, in order to identify the person's track in the video, it is necessary to first obtain a color picture containing the person's face.

[0053] S12. Obtain the face position data in the color picture containing the face image, and establish a fourth-order tensor model of the face of the color picture containing the face image according to the face position data;

[0054] It is understandable that since the fourth-order tensor model of the face is established based on the face position data in the face color picture, it not only contains more comprehensive face information but also weakens th...

Embodiment 2

[0062] Referring to embodiment one of the present invention and figure 1 The specific process of steps S11 to S16 described in , and refer to figure 2 Corresponding to Embodiment 2 of the present invention figure 1 A schematic flow diagram of establishing a fourth-order tensor model in step S12 shown in , the acquisition of the face position data in the color picture containing the face image, and the establishment of the face image containing the face position data according to the face position data The fourth-order tensor model of the face of the color picture, including:

[0063] S121. Convert the color picture containing the face image into a corresponding black and white picture, and extract the position data of the face in the black and white picture, wherein the position data includes initial coordinates, width data and height data;

[0064] Specifically, the main purpose of converting color pictures into black and white pictures to obtain location data is to reduce...

Embodiment 3

[0087] Corresponding to the method for identifying the trajectory of a person in a video disclosed in Embodiment 1 and Embodiment 2 of the present invention, Embodiment 3 of the present invention also provides a device for identifying the trajectory of a person in a video, see Figure 4 It is a schematic structural diagram of a device for identifying the trajectory of a person in a video provided by Embodiment 3 of the present invention. The device includes:

[0088] Conversion module 1 is used to convert the video to be identified into a plurality of color pictures comprising a face image according to whether the video segment contains a face image;

[0089] Build module 2, be used to obtain the face position data in the color picture that comprises human face image, set up the human face fourth-order tensor model of described color picture that comprises human face image according to described human face position data;

[0090] The first calculation module 3 is used to perf...

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 method and device for identifying character trajectory in a video. The method comprises the following steps: converting all video clips, comprising face images, in a video to be identified into a plurality of color pictures comprising the face images; obtaining face position data in the color pictures, and establishing a face fourth-order tensor module; carrying out tensor calculation and carrying out norm calculation and summation on a plurality of face feature tensor data and reference face tensor data; carrying out classification on faces in the color pictures correspondingly to obtain color images corresponding to the same face; and restoring the color images corresponding to the same face to the video to be identified to obtain the character trajectory of the face in the video to be identified. The method and device realize purposes of wide application range and reducing cost and calculation amount.

Description

technical field [0001] The invention relates to the technical field of face recognition, in particular to a tensor-based face recognition method and device. Background technique [0002] With the development of science and technology, traditional means of personal identity verification, such as passwords, certificates and IC cards, etc., due to the detachability of the holders, lead to forgery, embezzlement or deciphering, which cannot meet the needs of modern society and economy. activities and social security needs. Therefore, biometric identification technology emerged as the times require, and in biometric identification technology, because face recognition technology has the characteristics of convenient collection, easy acceptance and unfavorable forgery, face recognition technology has always been a research hotspot in biometric identification technology. [0003] The existing face recognition technology can be summarized as follows: After the face is detected and th...

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): G06K9/00
CPCG06V40/168G06V20/48G06V20/46
Inventor 徐佳宏李益永兰志才曾勇韩涛
Owner SHENZHEN IPANEL TECH LTD