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

The emergence of self-supervised learning in the field of wearables-basedhuman activity recognition (HAR) has opened up opportunities to tackle the mostpressing challenges in the field, namely to exploit unlabeled data to derivereliable recognition systems for scenarios where only small amounts of labeledtraining samples can be collected. As such, self-supervision, i.e., theparadigm of 'pretrain-then-finetune' has the potential to become a strongalternative to the predominant end-to-end training approaches, let alonehand-crafted features for the classic activity recognition chain. Recently anumber of contributions have been made that introduced self-supervised learninginto the field of HAR, including, Multi-task self-supervision, MaskedReconstruction, CPC, and SimCLR, to name but a few. With the initial success ofthese methods, the time has come for a systematic inventory and analysis of thepotential self-supervised learning has for the field. This paper providesexactly that. We assess the progress of self-supervised HAR research byintroducing a framework that performs a multi-faceted exploration of modelperformance. We organize the framework into three dimensions, each containingthree constituent criteria, such that each dimension captures specific aspectsof performance, including the robustness to differing source and targetconditions, the influence of dataset characteristics, and the feature spacecharacteristics. We utilize this framework to assess seven state-of-the-artself-supervised methods for HAR, leading to the formulation of insights intothe properties of these techniques and to establish their value towardslearning representations for diverse scenarios.

Assessing the State of Self-Supervised Human Activity Recognition using Wearables