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

In ambient assisted living, human activity recognition from time seriessensor data mainly focuses on predefined activities, often overlooking newactivity patterns. We propose CLAN, a two-tower contrastive learning-basednovelty detection framework with diverse types of negative pairs for humanactivity recognition. It is tailored to challenges with human activitycharacteristics, including the significance of temporal and frequency features,complex activity dynamics, shared features across activities, and sensormodality variations. The framework aims to construct invariant representationsof known activity robust to the challenges. To generate suitable negativepairs, it selects data augmentation methods according to the temporal andfrequency characteristics of each dataset. It derives the key representationsagainst meaningless dynamics by contrastive and classification losses-basedrepresentation learning and score function-based novelty detection thataccommodate dynamic numbers of the different types of augmented samples. Theproposed two-tower model extracts the representations in terms of time andfrequency, mutually enhancing expressiveness for distinguishing between new andknown activities, even when they share common features. Experiments on fourreal-world human activity datasets show that CLAN surpasses the bestperformance of existing novelty detection methods, improving by 8.3%, 13.7%,and 53.3% in AUROC, balanced accuracy, and FPR@TPR0.95 metrics respectively.

CLAN: A Contrastive Learning based Novelty Detection Framework for Human Activity Recognition