arXiv cs.AI· Zhengtao Gao, Francesco Bombassei De Bona, Leonardo Alchieri, Ivana Kiprijanovska, Simon Stankoski, Bojan Sofronievski, Hristijan Gjoreski, Marc Langheinrich, Martin Gjoreski·· 11 小时前AI 评分20
用自监督预训练实现 OCOsense 智能眼镜标签高效面部表情识别
Self-Supervised Pretraining for Label-Efficient Facial Expression Recognition using Smart Glasses
AI 导读
研究用掩码自编码器目标在无标注 OCOsense 智能眼镜记录上预训练 Vision Transformer，再微调做五类面部表情识别，在 5% 到 100% 各档标签预算下均优于从头训练。仅用 5% 标签时，预训练模型即超过用全部标签从头训练的模型。该工作获 UbiComp Companion '26 的 SEPH 2026 workshop 最佳论文奖。
正文 · 原文
Abstract:Facial Expression Recognition (FER) on optomyographic (OMG) smart glasses such as OCOsense enables privacy-preserving, camera-free affect sensing for health and well-being. Most FER systems are fully supervised, i.e., they rely either on hand-crafted expert features with classical machine learning, or on end-to-end deep models trained from labelled data. Both depend on large quantities of labelled expressions, which are costly to collect and limit deployment in real-world, personalized settings. We investigate Self-Supervised Learning (SSL) as a route to label-efficient FER on these glasses. We pretrain a Vision Transformer with a masked-autoencoder objective on unlabelled OCOsense recordings and finetune it for five-class FER under person-independent cross-validation. Across label budgets from 5% to 100%, pretraining consistently improves performance; with only 5% of the labels, the pretrained model outperforms the from-scratch model trained with all labels. Our results show that SSL pretraining is a practical way to reduce labelling cost for smart eyewear FER, and we discuss the implications for future personalization and continual adaptation.
| Comments: | 7 pages, 5 figures, 2 tables. Best Paper Award at the SEPH 2026 workshop, UbiComp Companion '26 |
| Subjects: | Artificial Intelligence (cs.AI) |
| ACM classes: | I.2.6; I.5.4 |
| Cite as: | arXiv:2611.05711 [cs.AI] |
| (or arXiv:2611.05711v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2611.05711 arXiv-issued DOI via DataCite (pending registration) |
|
| Journal reference: | Companion of the 2026 ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp Companion '26), October 11-15, 2026, Shanghai, China |
| Related DOI: | https://doi.org/10.1145/3798063.3839146
DOI(s) linking to related resources |
Submission history
From: Francesco Bombassei De Bona [view email]
[v1]
Tue, 3 Nov 2026 22:37:17 UTC (4,539 KB)
来源：arXiv cs.AI · arxiv.org