Salvato in:
Dettagli Bibliografici
Autori principali: Badatya, Bikash Kumar, Baghel, Vipul, Hegde, Ravi
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2508.19647
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866908505690079232
author Badatya, Bikash Kumar
Baghel, Vipul
Hegde, Ravi
author_facet Badatya, Bikash Kumar
Baghel, Vipul
Hegde, Ravi
contents Fine-grained action localization in untrimmed sports videos presents a significant challenge due to rapid and subtle motion transitions over short durations. Existing supervised and weakly supervised solutions often rely on extensive annotated datasets and high-capacity models, making them computationally intensive and less adaptable to real-world scenarios. In this work, we introduce a lightweight and unsupervised skeleton-based action localization pipeline that leverages spatio-temporal graph neural representations. Our approach pre-trains an Attention-based Spatio-Temporal Graph Convolutional Network (ASTGCN) on a pose-sequence denoising task with blockwise partitions, enabling it to learn intrinsic motion dynamics without any manual labeling. At inference, we define a novel Action Dynamics Metric (ADM), computed directly from low-dimensional ASTGCN embeddings, which detects motion boundaries by identifying inflection points in its curvature profile. Our method achieves a mean Average Precision (mAP) of 82.66% and average localization latency of 29.09 ms on the DSV Diving dataset, matching state-of-the-art supervised performance while maintaining computational efficiency. Furthermore, it generalizes robustly to unseen, in-the-wild diving footage without retraining, demonstrating its practical applicability for lightweight, real-time action analysis systems in embedded or dynamic environments.
format Preprint
id arxiv_https___arxiv_org_abs_2508_19647
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle UTAL-GNN: Unsupervised Temporal Action Localization using Graph Neural Networks
Badatya, Bikash Kumar
Baghel, Vipul
Hegde, Ravi
Computer Vision and Pattern Recognition
I.2.10; I.5.4
Fine-grained action localization in untrimmed sports videos presents a significant challenge due to rapid and subtle motion transitions over short durations. Existing supervised and weakly supervised solutions often rely on extensive annotated datasets and high-capacity models, making them computationally intensive and less adaptable to real-world scenarios. In this work, we introduce a lightweight and unsupervised skeleton-based action localization pipeline that leverages spatio-temporal graph neural representations. Our approach pre-trains an Attention-based Spatio-Temporal Graph Convolutional Network (ASTGCN) on a pose-sequence denoising task with blockwise partitions, enabling it to learn intrinsic motion dynamics without any manual labeling. At inference, we define a novel Action Dynamics Metric (ADM), computed directly from low-dimensional ASTGCN embeddings, which detects motion boundaries by identifying inflection points in its curvature profile. Our method achieves a mean Average Precision (mAP) of 82.66% and average localization latency of 29.09 ms on the DSV Diving dataset, matching state-of-the-art supervised performance while maintaining computational efficiency. Furthermore, it generalizes robustly to unseen, in-the-wild diving footage without retraining, demonstrating its practical applicability for lightweight, real-time action analysis systems in embedded or dynamic environments.
title UTAL-GNN: Unsupervised Temporal Action Localization using Graph Neural Networks
topic Computer Vision and Pattern Recognition
I.2.10; I.5.4
url https://arxiv.org/abs/2508.19647