A Topology-Aware Graph Convolutional Network for Human Pose Similarity and Action Quality Assessment

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
1. Verfasser: Zeng, Minmin
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866915592464760832
author Zeng, Minmin
author_facet Zeng, Minmin
contents Action Quality Assessment (AQA) requires fine-grained understanding of human motion and precise evaluation of pose similarity. This paper proposes a topology-aware Graph Convolutional Network (GCN) framework, termed GCN-PSN, which models the human skeleton as a graph to learn discriminative, topology-sensitive pose embeddings. Using a Siamese architecture trained with a contrastive regression objective, our method outperforms coordinate-based baselines and achieves competitive performance on AQA-7 and FineDiving benchmarks. Experimental results and ablation studies validate the effectiveness of leveraging skeletal topology for pose similarity and action quality assessment.
format Preprint
id arxiv_https___arxiv_org_abs_2511_01194
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A Topology-Aware Graph Convolutional Network for Human Pose Similarity and Action Quality Assessment
Zeng, Minmin
Computer Vision and Pattern Recognition
Artificial Intelligence
68T07 (Artificial neural networks and deep learning), 68U10 (Computer graphics, computational geometry)
Action Quality Assessment (AQA) requires fine-grained understanding of human motion and precise evaluation of pose similarity. This paper proposes a topology-aware Graph Convolutional Network (GCN) framework, termed GCN-PSN, which models the human skeleton as a graph to learn discriminative, topology-sensitive pose embeddings. Using a Siamese architecture trained with a contrastive regression objective, our method outperforms coordinate-based baselines and achieves competitive performance on AQA-7 and FineDiving benchmarks. Experimental results and ablation studies validate the effectiveness of leveraging skeletal topology for pose similarity and action quality assessment.
title A Topology-Aware Graph Convolutional Network for Human Pose Similarity and Action Quality Assessment
topic Computer Vision and Pattern Recognition
Artificial Intelligence
68T07 (Artificial neural networks and deep learning), 68U10 (Computer graphics, computational geometry)
url https://arxiv.org/abs/2511.01194