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

Semi-supervised facial motion unit detection method and system based on adversarial learning

A technology of motion unit and detection method, which is applied in the field of image recognition, can solve the problem of low detection method precision, achieve the effect of improving detection effect and enhancing robustness

Active Publication Date: 2021-12-10
HUAZHONG UNIV OF SCI & TECH
View PDF3 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] Aiming at the above defects or improvement needs of the prior art, the present invention provides a facial motion unit detection method and system, thereby solving the technical problem of low accuracy of the existing detection method

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
  • Semi-supervised facial motion unit detection method and system based on adversarial learning
  • Semi-supervised facial motion unit detection method and system based on adversarial learning
  • Semi-supervised facial motion unit detection method and system based on adversarial learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0032] In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not constitute a conflict with each other.

[0033] Embodiments of the present invention provide a semi-supervised facial motion unit detection method based on adversarial learning, such as figure 1 shown, including:

[0034] S1. Randomly select a source frame, a first target frame, and a second target frame from the training video, combine the optical flow vectors of the source frame and the first target frame with the source frame to ob...

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 semi-supervised facial motion unit detection method and system based on adversarial learning. The method comprises the steps: effectively capturing weak motion displacement through the full utilization of optical flow, namely, capturing the characteristics of AUs changes caused by facial expressions, and constructing a generative adversarial network model based on optical flow fusion for realizing accurate detection of a facial motion unit; in consideration of certain difficulty in selection of a vertex frame, setting the vertex frame and frames nearby the vertex frame as a vertex stage, randomly selecting a first target frame and a second target frame in the vertex stage, combining optical flow vectors of the first target frame and a source frame with the source frame to obtain a moving image as input of a generator, and taking the second target frame as a label to train the generator, so that the robustness of a model is enhanced, and the detection effect is further improved.

Description

technical field [0001] The invention belongs to the field of image recognition, and more particularly relates to a method and system for detecting facial motion units based on adversarial learning. Background technique [0002] In real-time video, facial expressions often occur simultaneously with head poses, and facial action units (ActionUnits, AUs) are the key factors for facial expression detection. The detection accuracy of existing facial action unit detection methods needs to be improved. Contents of the invention [0003] In view of the above defects or improvement needs of the prior art, the present invention provides a facial motion unit detection method and system, thereby solving the technical problem of low precision of the existing detection method. [0004] To achieve the above object, according to one aspect of the present invention, a method for detecting facial motion units based on adversarial learning is provided, including: [0005] S1, randomly selec...

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): G06K9/00G06N3/04G06N3/08
CPCG06N3/08G06N3/045
Inventor 喻莉杜聪炬
Owner HUAZHONG UNIV OF SCI & TECH