Salvato in:
Dettagli Bibliografici
Autori principali: Singh, Agamdeep, PB, Sujit, Vatsa, Mayank
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:https://arxiv.org/abs/2411.05734
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909380780228608
author Singh, Agamdeep
PB, Sujit
Vatsa, Mayank
author_facet Singh, Agamdeep
PB, Sujit
Vatsa, Mayank
contents Access to expert coaching is essential for developing technique in sports, yet economic barriers often place it out of reach for many enthusiasts. To bridge this gap, we introduce Poze, an innovative video processing framework that provides feedback on human motion, emulating the insights of a professional coach. Poze combines pose estimation with sequence comparison and is optimized to function effectively with minimal data. Poze surpasses state-of-the-art vision-language models in video question-answering frameworks, achieving 70% and 196% increase in accuracy over GPT4V and LLaVAv1.6 7b, respectively.
format Preprint
id arxiv_https___arxiv_org_abs_2411_05734
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Poze: Sports Technique Feedback under Data Constraints
Singh, Agamdeep
PB, Sujit
Vatsa, Mayank
Computer Vision and Pattern Recognition
Access to expert coaching is essential for developing technique in sports, yet economic barriers often place it out of reach for many enthusiasts. To bridge this gap, we introduce Poze, an innovative video processing framework that provides feedback on human motion, emulating the insights of a professional coach. Poze combines pose estimation with sequence comparison and is optimized to function effectively with minimal data. Poze surpasses state-of-the-art vision-language models in video question-answering frameworks, achieving 70% and 196% increase in accuracy over GPT4V and LLaVAv1.6 7b, respectively.
title Poze: Sports Technique Feedback under Data Constraints
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2411.05734