# Micro-expression Recognition Based on Dual-branch Feature Extraction and Fusion

arXiv: https://arxiv.org/abs/2602.23950 · HTML: https://mlsanity.com/paper/2602.23950/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 8 of its 19 references that are in the map.
- First posted on arXiv 27 Feb 2026.

## Builds on (most cited first, 8 of 8)

- [OFF-ApexNet on Micro-expression Recognition System](https://mlsanity.com/paper/1805.08699/) (2018)
- [Enriched Long-term Recurrent Convolutional Network for Facial Micro-Expression Recognition](https://mlsanity.com/paper/1805.08417/) (2018)
- [Feature refinement: An expression-specific feature learning and fusion method for micro-expression recognition](https://mlsanity.com/paper/2101.04838/) (2021)
- [LEARNet Dynamic Imaging Network for Micro Expression Recognition](https://mlsanity.com/paper/1904.09410/) (2019)
- [Short and Long Range Relation Based Spatio-Temporal Transformer for Micro-Expression Recognition](https://mlsanity.com/paper/2112.05851/) (2021)
- [Objective Classes for Micro-Facial Expression Recognition](https://mlsanity.com/paper/1708.07549/) (2017)
- [Recognizing Micro-Expression in Video Clip with Adaptive Key-Frame Mining](https://mlsanity.com/paper/2009.09179/) (2020)
- [Multi-scale multi-modal micro-expression recognition algorithm based on transformer](https://mlsanity.com/paper/2301.02969/) (2023)

## Abstract (arXiv)

Micro-expressions, characterized by transience and subtlety, pose challenges to existing optical flow-based recognition methods. To address this, this paper proposes a dual-branch micro-expression feature extraction network integrated with parallel attention. Key contributions include: 1) a residual network designed to alleviate gradient anishing and network degradation; 2) an Inception network constructed to enhance model representation and suppress interference from irrelevant regions; 3) an adaptive feature fusion module developed to integrate dual-branch features. Experiments on the CASME II dataset demonstrate that the proposed method achieves 74.67% accuracy, outperforming LBP-TOP (by 11.26%), MSMMT (by 3.36%), and other comparative methods.