.png){kind=icon}
Return
A gan inversion-based multimodal framework for micro-expression recognition
.png){kind=icon}
PRE
AI
DOI：10.1016/j.imavis.2026.105994.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• We propose a novel GAN Inversion-Based Multi-modal Framework(GANIM) for MER. • FD-KAN Module extracts identity-invariant expressions in the latent space. • HAUT and CMDA Modules effectively align and aggregate cross-modal features. • GANIM achieves state-of-the-art performance on SAMM, CASME2, CASME3.
Keywords:
GAN inversion
multi-modal framework
micro-expression recognition
feature alignment
latent space representation
Journal
IF:
4.2
Papers:
4.1K
Citations:
6.7K
Organization
Cited Papers
Dynamic Texture Recognition Using Local Binary Patterns with an Application to Facial Expressions
PREAI
ShareSave
ShareSaveSpontaneous facial micro-expression analysis using Spatiotemporal Completed Local Quantized Patterns.png)
NEUROCOMPUTING
NEUROCOMPUTINGIF6.5
PREAI
ShareSaveShareSaveFineCLIPER: Multi-modal Fine-grained CLIP for Dynamic Facial Expression Recognition with AdaptERs
PREAI
ShareSaveShareSaveShareSaveShareSave
View more