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

Various types of sensors have been considered to develop human actionrecognition (HAR) models. Robust HAR performance can be achieved by fusingmultimodal data acquired by different sensors. In this paper, we introduce anew multimodal fusion architecture, referred to as Unified Contrastive FusionTransformer (UCFFormer) designed to integrate data with diverse distributionsto enhance HAR performance. Based on the embedding features extracted from eachmodality, UCFFormer employs the Unified Transformer to capture theinter-dependency among embeddings in both time and modality domains. We presentthe Factorized Time-Modality Attention to perform self-attention efficientlyfor the Unified Transformer. UCFFormer also incorporates contrastive learningto reduce the discrepancy in feature distributions across various modalities,thus generating semantically aligned features for information fusion.Performance evaluation conducted on two popular datasets, UTD-MHAD and NTURGB+D, demonstrates that UCFFormer achieves state-of-the-art performance,outperforming competing methods by considerable margins.

Unified Contrastive Fusion Transformer for Multimodal Human Action Recognition