Enregistré dans:
Détails bibliographiques
Auteurs principaux: Nguyen, Son Hai, Wang, Diwei, Jang, Jinhyeok, Seo, Hyewon
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:https://arxiv.org/abs/2509.16452
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866918144827719680
author Nguyen, Son Hai
Wang, Diwei
Jang, Jinhyeok
Seo, Hyewon
author_facet Nguyen, Son Hai
Wang, Diwei
Jang, Jinhyeok
Seo, Hyewon
contents Accurate vision-based action recognition is crucial for developing autonomous robots that can operate safely and reliably in complex, real-world environments. In this work, we advance video-based recognition of indoor daily actions for robotic perception by leveraging vision-language models (VLMs) enriched with domain-specific knowledge. We adapt a prompt-learning framework in which class-level textual descriptions of each action are embedded as learnable prompts into a frozen pre-trained VLM backbone. Several strategies for structuring and encoding these textual descriptions are designed and evaluated. Experiments on the ETRI-Activity3D dataset demonstrate that our method, using only RGB video inputs at test time, achieves over 95\% accuracy and outperforms state-of-the-art approaches. These results highlight the effectiveness of knowledge-augmented prompts in enabling robust action recognition with minimal supervision.
format Preprint
id arxiv_https___arxiv_org_abs_2509_16452
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle KRAST: Knowledge-Augmented Robotic Action Recognition with Structured Text for Vision-Language Models
Nguyen, Son Hai
Wang, Diwei
Jang, Jinhyeok
Seo, Hyewon
Computer Vision and Pattern Recognition
Artificial Intelligence
Accurate vision-based action recognition is crucial for developing autonomous robots that can operate safely and reliably in complex, real-world environments. In this work, we advance video-based recognition of indoor daily actions for robotic perception by leveraging vision-language models (VLMs) enriched with domain-specific knowledge. We adapt a prompt-learning framework in which class-level textual descriptions of each action are embedded as learnable prompts into a frozen pre-trained VLM backbone. Several strategies for structuring and encoding these textual descriptions are designed and evaluated. Experiments on the ETRI-Activity3D dataset demonstrate that our method, using only RGB video inputs at test time, achieves over 95\% accuracy and outperforms state-of-the-art approaches. These results highlight the effectiveness of knowledge-augmented prompts in enabling robust action recognition with minimal supervision.
title KRAST: Knowledge-Augmented Robotic Action Recognition with Structured Text for Vision-Language Models
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2509.16452