Expression recognition method and system based on multi-scale features and spatial attention

By combining the HNFER neural network model with the CoordAtt and SAFM mechanisms, the difficulties of facial expression recognition in complex scenarios using traditional methods are solved, achieving highly accurate and robust facial expression recognition that is suitable for applications such as human-computer interaction and health diagnosis.

WO2025251557A1PCT designated stage Publication Date: 2025-12-11YANTAI UNIV
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
PCT/CN2024/135203
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2024-06-04
Filing Date
2024-11-28
Publication Date
2025-12-11

AI Technical Summary

Technical Problem

Existing technologies are ineffective in handling complex scenarios such as varying lighting, facial expression intensity, and facial occlusion. Traditional facial expression recognition methods struggle to accurately capture and process subtle changes in expression and lack dynamic attention to local features.

Method used

An expression recognition method based on multi-scale features and spatial attention is adopted. Feature extraction is performed through the HNFER neural network model, and the CoordAtt and SAFM mechanisms are combined to enhance the feature extraction and classification capabilities, including multi-scale feature extraction, spatial attention mechanism and deep convolution processing.

Benefits of technology

It improves the accuracy and robustness of facial expression recognition, maintains high performance in complex environments, and is suitable for fields such as human-computer interaction, security systems, and health diagnosis.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN2024135203_11122025_PF_FP_ABST
    Figure CN2024135203_11122025_PF_FP_ABST
Patent Text Reader

Abstract

The present application relates to the technical field of expression recognition, and in particular to an expression recognition method and system based on multi-scale features and spatial attention. The method comprises: using an HNFER neural network model to perform feature extraction on acquired facial image data to obtain an original input feature map; performing pooling and concatenation on the extracted features on the basis of a CoordAtt attention mechanism to obtain a feature map; performing deep convolution processing on the feature map to obtain an attention map, and then obtaining a final feature map by means of element multiplication; and performing feature transformation and normalization on the final feature map to obtain expression category probabilities and outputting the expression category probabilities . In the present application, by integrating scale-aware technology and spatial attention technology, a model can more accurately recognize and categorize different emotional states, and can maintain high performance even under complex environmental conditions.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Multi-scale feature extraction and attention mechanism fused expression recognition method

    CN115909455A

  • Facial expression recognition method based on space attention network

    CN116092146A

  • Facial expression recognition method and system based on feature fusion and attention mechanism

    CN116189272A

  • Multi-person real-time facial expression recognition method and system based on attention mechanism

    CN117636436A

  • Expression recognition method and system based on multi-scale features and space attention

    CN118298491A