Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Benavent-Lledo, Manuel, Mulero-Pérez, David, Ortiz-Perez, David, Garcia-Rodriguez, Jose
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2501.13518
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866911406385790976
author Benavent-Lledo, Manuel
Mulero-Pérez, David
Ortiz-Perez, David
Garcia-Rodriguez, Jose
author_facet Benavent-Lledo, Manuel
Mulero-Pérez, David
Ortiz-Perez, David
Garcia-Rodriguez, Jose
contents Detecting actions as they occur is essential for applications like video surveillance, autonomous driving, and human-robot interaction. Known as online action detection, this task requires classifying actions in streaming videos, handling background noise, and coping with incomplete actions. Transformer architectures are the current state-of-the-art, yet the potential of recent advancements in computer vision, particularly vision-language models (VLMs), remains largely untapped for this problem, partly due to high computational costs. In this paper, we introduce TOAD: a Text-driven Online Action Detection architecture that supports zero-shot and few-shot learning. TOAD leverages CLIP (Contrastive Language-Image Pretraining) textual embeddings, enabling efficient use of VLMs without significant computational overhead. Our model achieves 82.46% mAP on the THUMOS14 dataset, outperforming existing methods, and sets new baselines for zero-shot and few-shot performance on the THUMOS14 and TVSeries datasets.
format Preprint
id arxiv_https___arxiv_org_abs_2501_13518
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Text-driven Online Action Detection
Benavent-Lledo, Manuel
Mulero-Pérez, David
Ortiz-Perez, David
Garcia-Rodriguez, Jose
Computer Vision and Pattern Recognition
Detecting actions as they occur is essential for applications like video surveillance, autonomous driving, and human-robot interaction. Known as online action detection, this task requires classifying actions in streaming videos, handling background noise, and coping with incomplete actions. Transformer architectures are the current state-of-the-art, yet the potential of recent advancements in computer vision, particularly vision-language models (VLMs), remains largely untapped for this problem, partly due to high computational costs. In this paper, we introduce TOAD: a Text-driven Online Action Detection architecture that supports zero-shot and few-shot learning. TOAD leverages CLIP (Contrastive Language-Image Pretraining) textual embeddings, enabling efficient use of VLMs without significant computational overhead. Our model achieves 82.46% mAP on the THUMOS14 dataset, outperforming existing methods, and sets new baselines for zero-shot and few-shot performance on the THUMOS14 and TVSeries datasets.
title Text-driven Online Action Detection
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2501.13518