Use method of head-mounted expression capture device based on facial muscle characteristics and enhancement method of expression data packet

Abstract

The invention discloses a use method of a head-mounted expression capture device based on human face muscle features, which comprises the following steps: S1, adhering a multi-directional tension sensor to feature points of a face, and adhering circumferential fibers and central axis fibers along the contour of the face; s2, connecting the adjacent multidirectional tension sensors or the circumferential fibers and the middle shaft fibers by using tension fibers to ensure that the multidirectional tension sensors are connected with the tension fibers in at least three directions; s3, wearing the tension band and the image acquisition rod connected with the tension band on the top of the head; s4, when the face makes an expression, enabling the high-speed photographic device in the image acquisition rod to capture the facial expression, and enabling the multidirectional tension sensor at each position to capture a corresponding numerical value; and S5, packaging the data statistical graph of each position and the acquired facial expression static image into an expression data packet. Muscle movement data statistical graphs and expression static images, captured by the device, of thehuman face are packaged into an expression data packet, and convenience is provided for later movie and television special effects.

Description

technical field [0001] The invention relates to the technical field of feature recognition, in particular to a method for using a head-mounted expression capture device based on facial muscle features and an enhancement method for expression data packets. Background technique [0002] The facial expression of a human face is an important feature for people to express feelings or emotions, and the analysis of facial expressions can effectively obtain people's emotional information. The collection of expressions has become an important part of expression recognition and output, and effective collection of expressions is the basis of emotion recognition. Traditional face collection and expression recognition is mainly realized by extracting local features or overall features of the face image. This collection method only considers the surface contour information of the face, and cannot describe the muscle movement of the face when making an expression. Minor changes, so the ac...

AI Technical Summary

Problems solved by technology

Traditional face collection and expression recognition is mainly realized by extracting local features or overall features of the face image. This collection method only considers the surface contour information of the face, and cannot describe the muscle movement of the face when making an expression. Small changes, so the accuracy of facial expression collection and recognition is not high

[0003] The traditional face expression data package only has static pictures of different expressions. Most of the enhancement methods for each picture are made up by big data, which requires a long time to accumulate the number of samples

Images