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

A multi-modal personalized emotion processing method based on long-short-term memory mechanism

A technology of long-short-term memory and processing methods, applied in neural learning methods, special data processing applications, computer components, etc., can solve problems such as lack of emotional uncertainty, failure to consider influence, and neglect of relevance

Active Publication Date: 2022-03-22
SHANTOU UNIV
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, in these emotion studies, there are generally the following three problems: (1) For emotion cognition, researchers pay more attention to the classification accuracy, but lack of consideration of the uncertainty of emotion
(2) For emotion understanding, few people have studied it, especially in multimedia video content; (3) For emotion expression, since most simulation experiments do not use real "stimuli", this will lead to a lack of simulation effect good practical sense
[0007] 1. The classification method of the classification results with the priority of accuracy rate is not reasonable, especially in complex situations such as uncertainty and emotion
[0008] 2. When studying emotion classification or emotion calculation, visual neurons, auditory neurons and some neurons generated by experience in the brain are ignored. The importance of the combination of these three neurons, so the final classification effect of the existing model, It's all one-sided
[0009] 3. The influence of the emotion of the "stimulus" on the emotion transfer matrix is ​​not considered, that is, the relationship between the external "stimulus" and the emotion of the human brain is not established
However, current researchers do not view emotional decision-making from this objective perspective.
[0012] 2. Current researchers often study emotion models in a single modality, often ignoring the correlation with other modality data, resulting in a one-sided cognitive effect
[0013] 3. Although the input of "stimuli" is one of the key influencing factors of the state transition matrix, there are few studies at present
In addition, a small number of researchers who study the emotion of "stimuli" have not proposed a personality-emotional state transition model suitable for real fuzzy stimulus input, nor have they considered the input of "stimuli" as an equivalent problem-emotion classification. solve

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 multi-modal personalized emotion processing method based on long-short-term memory mechanism
  • A multi-modal personalized emotion processing method based on long-short-term memory mechanism
  • A multi-modal personalized emotion processing method based on long-short-term memory mechanism

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0080] 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.

[0081] Such as figure 1 , the multi-modal personalized emotion processing method based on the long-short-term memory mechanism mainly includes: information preprocessing and extraction, emotion recognition unit, and emotion calculation unit.

[0082] The preprocessing and extraction of information includes the following steps:

[0083] S110. Information collection, that is, collection of Music video (music video) data.

[0084] S120. Information preprocessing.

[0085] S130, feature extraction.

[0086] In S110, 30 Chinese music videos were selected, 15 of which expressed positive emotions (inspiring, warm, romantic, happy, etc.), and 15 expressed negative themes (broken love, homesickness, frustration, etc.). Then, 30 music soundtracks were cut into 498 MSUs ...

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 embodiment of the present invention discloses a multi-modal personalized emotion processing method based on the long-short-term memory mechanism. Taking music videos as "stimuli", a multi-modal personalized processing method based on the long-short-term memory mechanism is designed to automatically perceive And extract the characteristics of "stimuli", understand and remember the stimuli, and then simulate the process of human emotion transfer to realize the expression of emotion. The invention belongs to the field of artificial intelligence and relates to an artificial emotion simulation system, which can be applied to the development of emotional and intelligent new human-computer interaction products. In addition, starting from the brain-like mechanism, the present invention can provide a reference idea for simulating, extending and expanding the natural brain-like emotion of intelligent robots.

Description

technical field [0001] The invention relates to the field of artificial intelligence, in particular to an artificial emotion simulation system, which can be applied to the development of an emotional and intelligent new human-computer interaction method. Background technique [0002] Emotion is considered to be the final standard that distinguishes human behavior from all other biological behaviors. To make computers have higher and more comprehensive intelligence, computers must be endowed with the ability to recognize, understand, express and adapt to human emotions to establish a harmonious human being. machine environment. This is the ultimate goal in the field of affective computing, and it is also the only way to go from weak artificial intelligence to strong artificial intelligence. With the continuous deepening of research, the research content gradually expands and extends. People are no longer limited to mechanically calculating emotions, but systematic research o...

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): G06F16/75G06F16/78G06V10/774G06V10/764G06V10/82G06K9/62G06N3/04G06N3/08
CPCG06F16/75G06F16/7867G06N3/08G06N3/044G06N3/045G06F18/241G06F18/214
Inventor 姜大志涂耿
Owner SHANTOU 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