From Macro to Micro: Boosting micro-expression recognition via pre-training on macro-expression videos

TL;DR

This paper introduces a transfer learning approach that leverages macro-expression videos to improve micro-expression recognition, addressing data scarcity and enhancing subtle facial movement detection.

Contribution

It proposes a novel MA2MI transfer learning paradigm and a two-branch network (MIACNet) to better capture facial action features for MER.

Findings

Outperforms existing MER methods on benchmark datasets.

Effectively utilizes macro-expression data to enhance micro-expression recognition.

Improves localization of facial action regions.

Abstract

Micro-expression recognition (MER) has drawn increasing attention in recent years due to its potential applications in intelligent medical and lie detection. However, the shortage of annotated data has been the major obstacle to further improve deep-learning based MER methods. Intuitively, utilizing sufficient macro-expression data to promote MER performance seems to be a feasible solution. However, the facial patterns of macro-expressions and micro-expressions are significantly different, which makes naive transfer learning methods difficult to deploy directly. To tacle this issue, we propose a generalized transfer learning paradigm, called \textbf{MA}cro-expression \textbf{TO} \textbf{MI}cro-expression (MA2MI). Under our paradigm, networks can learns the ability to represent subtle facial movement by reconstructing future frames. In addition, we also propose a two-branch micro-action network (MIACNet) to decouple facial position features and facial action features, which can help the network more accurately locate facial action locations. Extensive experiments on three popular MER benchmarks demonstrate the superiority of our method.