MagPlus: Bridging Micro-to-Regular Facial Expressions through Learnable Magnification

2026-06-11

AI Summary

Micro-expression modeling faces significant challenges due to data scarcity and subtle motion dynamics, often resulting in low generation quality, poor robustness, and limited generalization. To address these issues, this work proposes the MagPlus framework, which introduces a learnable motion amplification mechanism (MagPlus) to map micro-expressions into regular facial expression signals. This enables direct transfer synthesis using state-of-the-art macro-expression animation models—such as FOMM and FSRT—without requiring any retraining. A subsequent inverse restoration module (DeMagPlus) then recovers the amplified expressions back to their original, realistic intensity levels. Notably, MagPlus achieves high-fidelity micro-expression generation without any exposure to micro-expression data during training, substantially enhancing both the realism and dynamic expressiveness of the synthesized results.

Abstract

Facial micro-expressions are subtle and short-lived facial movements that provide important cues about genuine human emotions. However, modeling and generating them remains difficult because annotated micro-expression data is limited and the underlying facial motions are extremely weak. Existing micro-expression generation methods therefore often suffer from limited quality, weak robustness, and poor generalization. We propose MagPlus, a transferable micro-expression processing pipeline that connects micro-expression analysis with standard facial animation models. Instead of training a dedicated generator from scratch, MagPlus learns to magnify subtle facial motions into the range of regular facial expressions, transforming micro-expressions into signals that are compatible with existing facial expression processing models. The magnified sequence is then used by a standard facial expression model for tasks such as transfer and synthesis. A complementary DeMagPlus module then restores the generated motion back to realistic micro-expression intensity levels while preserving the synthesized dynamics. We evaluate the framework using four facial animation models: FOMM, FSRT, MetaPortrait, and EmoPortraits. None of these models are trained on micro-expression data. Experiments show that MagPlus-DeMagPlus enables pretrained macro-expression models to generate more realistic micro-expression motion without retraining the backbones.

Problem

• micro-expression
• facial animation
• motion magnification
• expression generation
• data scarcity

Innovation

• micro-expression magnification
• transferable facial animation
• learnable motion enhancement
• DeMagPlus
• macro-to-micro expression bridging