Digital human training data cleaning method, device, electronic equipment and storage medium

A training data and digital human technology, applied in the field of image processing, can solve problems such as high time and labor costs, low detection efficiency, and training data that do not meet the standards, so as to avoid labor costs and time costs, effectively clean and improve The effect of cleaning efficiency

Pending Publication Date: 2020-11-20
SHENZHEN ZHUIYI TECH CO LTD
View PDF5 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] However, the training data obtained through real-person recording videos may not meet the standards. At present, the detection of

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
  • Digital human training data cleaning method, device, electronic equipment and storage medium
  • Digital human training data cleaning method, device, electronic equipment and storage medium
  • Digital human training data cleaning method, device, electronic equipment and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Example

[0048] first embodiment

[0049] see figure 1 , figure 1 A schematic diagram of an application environment suitable for the embodiments of the present application is shown. The method for cleaning digital human training data provided by the embodiments of the present application should be used for example figure 1 Interactive system 100 is shown. The interactive system 100 includes a terminal device 101 and a server 102 , and the server 102 is connected to the terminal device 101 in communication. The server 102 may be a traditional server or a cloud server, which is not specifically limited herein.

[0050] The terminal device 101 may be various electronic devices with a display screen, a data processing module, a camera, audio input / output and other functions, and supports data input, including but not limited to smart phones, tablet computers, laptops portable computers, desktop computers, self-service terminals and wearable electronic devices, etc. Specifically, the ...

Example

[0068] Second Embodiment

[0069] see image 3 , image 3 A schematic flowchart of the method for cleaning digital human training data provided by the embodiment of the present application is shown. The method may include the following steps:

[0070] S210: Acquire the to-be-processed segment in the digital human training data, and the to-be-processed segment records and generates target line information based on the target object.

[0071] S220, obtain standard character feature information based on the target line information.

[0072] S230: Acquire characteristic information of the person under test of the target object from the segment to be processed.

[0073] S240: Determine whether the characteristic information of the person to be measured matches the standard characteristic information of the person.

[0074] The specific implementation manners of S210 to S240 can be referred to S110 to S140, so they are not repeated here.

[0075] S250 , if the characteristic i...

Example

[0084] Third Embodiment

[0085] see Figure 5 , Figure 5 A schematic flowchart of the method for cleaning digital human training data provided by the embodiment of the present application is shown. The method may include the following steps:

[0086] S310: Acquire the to-be-processed segment in the digital human training data, and the to-be-processed segment records and generates target line information based on the target object.

[0087] S320, obtain standard character feature information based on the target line information.

[0088] S330: Acquire character feature information of the target object to be measured from the segment to be processed.

[0089] The specific implementation manners of S310 to S330 can be referred to S110 to S130, so they are not repeated here.

[0090] S340, determine whether there is an image including an abnormal object in the segment to be processed.

[0091] In some embodiments, a standard image may be preset, and the standard image may ...

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 digital human training data cleaning method, a device, electronic equipment and a storage medium, and relates to the technical field of image processing. The method comprisesthe steps of obtaining a to-be-processed segment in digital human training data, and the to-be-processed segment being generated by recording target line information based on a target object; collecting standard character feature information based on the target line information; obtaining to-be-tested character feature information of the target object from the to-be-processed fragment; judging whether the to-be-detected character feature information is matched with the standard character feature information or not; if the to-be-detected character feature information is not matched with the standard character feature information, deleting the to-be-processed fragment. According to the invention, the cleaning efficiency and the cleaning quality of the training data can be improved.

Description

technical field [0001] The present application relates to the technical field of image processing, and more specifically, to a method, device, electronic equipment and storage medium for cleaning digital human training data. Background technique [0002] With the rapid development of science and technology, human-computer interaction technology has penetrated into all aspects of daily life. Among them, digital virtual humans are becoming more and more important in human-computer interaction technology. Digital virtual humans use virtual reality technology, human-computer interaction, A virtual 3D person produced by high-precision 3D portrait simulation, artificial intelligence (AI), motion capture, facial expression capture and other technologies. Among them, when making a digital avatar, the training data is usually obtained by recording a video of a real person, and then the training data is trained with the digital avatar, so that the digital avatar can move like a real p...

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): G06F16/215G06F16/583
CPCG06F16/215G06F16/583G06F16/5854
Inventor 常向月
Owner SHENZHEN ZHUIYI TECH CO LTD
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