English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Cutting-edge research in facial expression recognition (FER) currently favorsthe utilization of convolutional neural networks (CNNs) backbone which issupervisedly pre-trained on face recognition datasets for feature extraction.However, due to the vast scale of face recognition datasets and the high costassociated with collecting facial labels, this pre-training paradigm incurssignificant expenses. Towards this end, we propose to pre-train visionTransformers (ViTs) through a self-supervised approach on a mid-scale generalimage dataset. In addition, when compared with the domain disparity existingbetween face datasets and FER datasets, the divergence between general datasetsand FER datasets is more pronounced. Therefore, we propose a contrastivefine-tuning approach to effectively mitigate this domain disparity.Specifically, we introduce a novel FER training paradigm named Mask Imagepre-training with MIx Contrastive fine-tuning (MIMIC). In the initial phase, wepre-train the ViT via masked image reconstruction on general images.Subsequently, in the fine-tuning stage, we introduce a mix-supervisedcontrastive learning process, which enhances the model with a more extensiverange of positive samples by the mixing strategy. Through extensive experimentsconducted on three benchmark datasets, we demonstrate that our MIMICoutperforms the previous training paradigm, showing its capability to learnbetter representations. Remarkably, the results indicate that the vanilla ViTcan achieve impressive performance without the need for intricate,auxiliary-designed modules. Moreover, when scaling up the model size, MIMICexhibits no performance saturation and is superior to the currentstate-of-the-art methods.

MIMIC: Mask Image Pre-training with Mix Contrastive Fine-tuning for Facial Expression Recognition