GT-SRR: A Structured Method for Social Relation Recognition with GGNN-Based Transformer

Social relationship recognition (SRR) holds significant value in fields such as behavior analysis and intelligent social systems. However, existing methods primarily focus on modeling individual visual traits, interaction patterns, and scene-level contextual cues, often failing to capture the comple...

Full description

Saved in:
Bibliographic Details
Main Authors: Dejiao Huang, Menglei Xia, Ruyi Chang, Xiaohan Kong, Shuai Guo
Format: Article
Language:English
Published: MDPI AG 2025-05-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/25/10/2992
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849718945063370752
author Dejiao Huang
Menglei Xia
Ruyi Chang
Xiaohan Kong
Shuai Guo
author_facet Dejiao Huang
Menglei Xia
Ruyi Chang
Xiaohan Kong
Shuai Guo
author_sort Dejiao Huang
collection DOAJ
description Social relationship recognition (SRR) holds significant value in fields such as behavior analysis and intelligent social systems. However, existing methods primarily focus on modeling individual visual traits, interaction patterns, and scene-level contextual cues, often failing to capture the complex dependencies among these features and the hierarchical structure of social groups, which are crucial for effective reasoning. In order to overcome these restrictions, this essay suggests a SRR model that integrates Gated Graph Neural Network (GGNN) and Transformer. The task for SRR in this model is image-based. Specifically, the purpose of a novel and robust hybrid feature extraction module is to capture individual characteristics, relative positional information, and group-level cues, which are used to construct relation nodes and group nodes. A modified GGNN is then employed to model the logical dependencies between features. Nevertheless, GGNN alone lacks the capacity to dynamically adjust feature importance, which may result in ambiguous relationship representations. The Transformer’s multi-head self-attention (MSA) mechanism is integrated to improve feature interaction modeling, allowing the model to capture global context and higher-order dependencies effectively. By fusing pairwise features, graph-structured features, and group-level information. Experimental results on public datasets such as PISC demonstrate that the proposed approach outperforms comparison models including Dual-Glance, GRM, GRRN, Graph-BERT, and SRT in terms of accuracy and mean average precision (mAP), validating its effectiveness in multi-feature representation learning and global reasoning.
format Article
id doaj-art-25dbc14f96b646d88548542e4eda7269
institution DOAJ
issn 1424-8220
language English
publishDate 2025-05-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj-art-25dbc14f96b646d88548542e4eda72692025-08-20T03:12:15ZengMDPI AGSensors1424-82202025-05-012510299210.3390/s25102992GT-SRR: A Structured Method for Social Relation Recognition with GGNN-Based TransformerDejiao Huang0Menglei Xia1Ruyi Chang2Xiaohan Kong3Shuai Guo4College of Science, Qingdao University of Technology, Qingdao 266000, ChinaCollege of Information and Control Engineering, Qingdao University of Technology, Qingdao 266000, ChinaCollege of Science, Qingdao University of Technology, Qingdao 266000, ChinaCollege of Science, Qingdao University of Technology, Qingdao 266000, ChinaCollege of Information and Control Engineering, Qingdao University of Technology, Qingdao 266000, ChinaSocial relationship recognition (SRR) holds significant value in fields such as behavior analysis and intelligent social systems. However, existing methods primarily focus on modeling individual visual traits, interaction patterns, and scene-level contextual cues, often failing to capture the complex dependencies among these features and the hierarchical structure of social groups, which are crucial for effective reasoning. In order to overcome these restrictions, this essay suggests a SRR model that integrates Gated Graph Neural Network (GGNN) and Transformer. The task for SRR in this model is image-based. Specifically, the purpose of a novel and robust hybrid feature extraction module is to capture individual characteristics, relative positional information, and group-level cues, which are used to construct relation nodes and group nodes. A modified GGNN is then employed to model the logical dependencies between features. Nevertheless, GGNN alone lacks the capacity to dynamically adjust feature importance, which may result in ambiguous relationship representations. The Transformer’s multi-head self-attention (MSA) mechanism is integrated to improve feature interaction modeling, allowing the model to capture global context and higher-order dependencies effectively. By fusing pairwise features, graph-structured features, and group-level information. Experimental results on public datasets such as PISC demonstrate that the proposed approach outperforms comparison models including Dual-Glance, GRM, GRRN, Graph-BERT, and SRT in terms of accuracy and mean average precision (mAP), validating its effectiveness in multi-feature representation learning and global reasoning.https://www.mdpi.com/1424-8220/25/10/2992SRRGGNNtransformerfeature interaction modeling
spellingShingle Dejiao Huang
Menglei Xia
Ruyi Chang
Xiaohan Kong
Shuai Guo
GT-SRR: A Structured Method for Social Relation Recognition with GGNN-Based Transformer
Sensors
SRR
GGNN
transformer
feature interaction modeling
title GT-SRR: A Structured Method for Social Relation Recognition with GGNN-Based Transformer
title_full GT-SRR: A Structured Method for Social Relation Recognition with GGNN-Based Transformer
title_fullStr GT-SRR: A Structured Method for Social Relation Recognition with GGNN-Based Transformer
title_full_unstemmed GT-SRR: A Structured Method for Social Relation Recognition with GGNN-Based Transformer
title_short GT-SRR: A Structured Method for Social Relation Recognition with GGNN-Based Transformer
title_sort gt srr a structured method for social relation recognition with ggnn based transformer
topic SRR
GGNN
transformer
feature interaction modeling
url https://www.mdpi.com/1424-8220/25/10/2992
work_keys_str_mv AT dejiaohuang gtsrrastructuredmethodforsocialrelationrecognitionwithggnnbasedtransformer
AT mengleixia gtsrrastructuredmethodforsocialrelationrecognitionwithggnnbasedtransformer
AT ruyichang gtsrrastructuredmethodforsocialrelationrecognitionwithggnnbasedtransformer
AT xiaohankong gtsrrastructuredmethodforsocialrelationrecognitionwithggnnbasedtransformer
AT shuaiguo gtsrrastructuredmethodforsocialrelationrecognitionwithggnnbasedtransformer