Micro-expression recognition method and device based on multi-scale optical flow attention guiding mechanism

The micro-expression recognition method using a multi-scale optical flow-guided attention mechanism solves the challenges of capturing subtle features and dynamic temporal modeling in micro-expression recognition by utilizing Transformer blocks and a dual-path global attention mechanism, thereby improving recognition accuracy and robustness.

CN121661698AActive Publication Date: 2026-03-13CHINA JILIANG UNIV
4 Cites 3 Cited by

Patent Information

Application Number
CN202610171075.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-02-06
Publication Date
2026-03-13
Estimated Expiration
2046-02-06

AI Technical Summary

Technical Problem

Existing micro-expression recognition technologies struggle to effectively capture subtle features, and dynamic temporal modeling is complex and data is scarce, resulting in limited model generalization ability. Furthermore, insufficient capture of cross-modal temporal correlation and frequency domain information affects recognition accuracy.

Method used

A multi-scale optical flow-guided attention mechanism is adopted. By constructing a pre-trained model and combining dynamic motion branch, spatiotemporal appearance branch and cross-modal fusion module, multi-scale optical flow features and temporal difference features are extracted using Transformer blocks. Feature fusion is then performed using a dual-path global attention mechanism to improve the sensitivity to micro-motions in key regions.

Benefits of technology

It significantly improves the accuracy of micro-expression recognition, especially the recognition performance on different datasets, solves the problems of capturing subtle features and dynamic temporal modeling, and enhances the robustness and recognition accuracy of the model.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention provides a micro-expression recognition method and device based on a multi-scale optical flow guiding attention mechanism, and the method comprises the following steps: constructing a pre-trained micro-expression recognition model which comprises a preprocessing module, a dynamic motion branch and a space-time appearance branch which are arranged in parallel, a cross-modal fusion module, and a classification head; and obtaining a to-be-recognized micro-expression video, and inputting the to-be-recognized micro-expression video into the micro-expression recognition model to obtain an emotion category According to the scheme, the attention sensitivity of the model to micro-motion of key areas such as orbiculus oculi muscle and zygomatic lesser muscle is remarkably improved under the guidance of the optical flow dynamic information, so that the accuracy of emotion category prediction is improved.
Need to check novelty before this filing date? Find Prior Art