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

A method and device for active face finding using local observation information

An active face, local technology, applied in the field of active face finding, can solve problems such as application obstacles

Active Publication Date: 2020-06-02
TSINGHUA UNIV
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

caused some application barriers

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 method and device for active face finding using local observation information
  • A method and device for active face finding using local observation information
  • A method and device for active face finding using local observation information

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0076] A method and device for active face finding using local observation information proposed by the present invention, combined with the accompanying drawings and embodiments, are described in detail as follows:

[0077] The present invention proposes an active face finding method using partial observation information, the partial observation information refers to the observation information does not contain or contains part of the face, the method is divided into two stages: the training stage and the actual execution stage: training In the stage, the training data collected offline is trained to obtain the neural network and its parameters and saved. In the actual execution stage, the parameters of the neural network obtained in the training stage are loaded and actions are performed according to the neural network. Include the following steps:

[0078] 1) Training stage: the training stage is to conduct offline training on the basis of the collected training data to obta...

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 an active face finding method and device using local observation information, and belongs to the field of image processing. The device includes: camera, execution device, bracket, connecting line and computing unit. In the training phase, the method collects images of the same environment with different setting scenes to form a training scene set, randomly selects a training scene in the training scene set, uses a neural network to generate the action to be executed, and simulates the actual execution of the camera to obtain a trajectory of action execution. The trajectory is evaluated and optimized, and the optimized neural network parameters are obtained, and the parameters are judged to obtain the final neural network parameters; in the actual execution stage, the camera captures the picture in real time and detects whether there is a face in the current picture. End; if it does not exist, the action should be executed through the neural network calculation, and the action is re-determined after the action is executed. The invention effectively improves the utilization efficiency of data, and has better adaptability and generalization ability.

Description

technical field [0001] The invention belongs to the field of image processing, in particular to an active face finding method and device using local observation information. Background technique [0002] Active face detection refers to the process in which the computing unit calculates the action command through the information obtained by the current camera observation, and sends it to the execution device to make the camera move within the movable range and actively search for the face. As a technical difficulty of service robots and video intercom robots, it plays a vital role in improving the quality of natural interaction between robots and humans. Especially for chatbots, the main audience is the elderly. When the elderly use the video intercom function, because it is inconvenient to move, sometimes the camera cannot be aligned with the face, causing certain communication obstacles. Therefore, an effective face-finding solution is needed to enable the machine to autom...

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): G06K9/00G06N3/08G05D3/12
CPCG05D3/12G06N3/08G06V40/161
Inventor 刘华平刘昌祥吴雨培孙富春
Owner TSINGHUA 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

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com