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

Vision-based human activity recognition (HAR) has made substantial progressin recognizing predefined gestures but lacks adaptability for emergingactivities. This paper introduces a paradigm shift by harnessing generativemodeling and large language models (LLMs) to enhance vision-based HAR. Wepropose utilizing LLMs to generate descriptive textual representations ofactivities using pose keypoints as an intermediate representation.Incorporating pose keypoints adds contextual depth to the recognition process,allowing for sequences of vectors resembling text chunks, compatible with LLMs.This innovative fusion of computer vision and natural language processing holdssignificant potential for revolutionizing activity recognition. A proof ofconcept study on a Kinetics700 dataset subset validates the approach'sefficacy, highlighting improved accuracy and interpretability. Futureimplications encompass enhanced accuracy, novel research avenues, modelgeneralization, and ethical considerations for transparency. This framework hasreal-world applications, including personalized gym workout feedback andnuanced sports training insights. By connecting visual cues to interpretabletextual descriptions, the proposed framework advances HAR accuracy andapplicability, shaping the landscape of pervasive computing and activityrecognition research. As this approach evolves, it promises a more insightfulunderstanding of human activities across diverse contexts, marking asignificant step towards a better world.

Towards Enhanced Human Activity Recognition through Natural Language Generation and Pose Estimation