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

Efficient human activity recognition (HAR) using sensor data needs asignificant volume of annotated data. The growing volume of unlabelled sensordata has challenged conventional practices for gathering HAR annotations withhuman-in-the-loop approaches, often leading to the collection of shallowerannotations. These shallower annotations ignore the fine-grainedmicro-activities that constitute any complex activities of daily living (ADL).Understanding this, we, in this paper, first analyze this lack of granularannotations from available pre-annotated datasets to understand the practicalinconsistencies and also perform a detailed survey to look into the humanperception surrounding annotations. Drawing motivations from these, we nextdevelop the framework AmicroN that can automatically generate micro-activityannotations using locomotive signatures and the available coarse-grainmacro-activity labels. In the backend, AmicroN applies change-point detectionfollowed by zero-shot learning with activity embeddings to identify the unseenmicro-activities in an unsupervised manner. Rigorous evaluation on publiclyavailable datasets shows that AmicroN can accurately generate micro-activityannotations with a median F1-score of >0.75. Additionally, we also show thatAmicroN can be used in a plug-and-play manner with Large Language Models (LLMs)to obtain the micro-activity labels, thus making it more practical forrealistic applications.

AmicroN: A Framework for Generating Annotations for Human Activity Recognition with Granular Micro-Activities