DualCascadeTSF-MobileNetV2: A Lightweight Violence Behavior Recognition Model

This paper proposes a lightweight violent behavior recognition model, DualCascadeTSF-MobileNetV2, which is improved based on the temporal shift module and its subsequent research. By introducing the Dual Cascade Temporal Shift and Fusion module, the model further enhances the feature correlation abi...

Full description

Saved in:
Bibliographic Details
Main Authors: Yuang Chen, Yong Li, Shaohua Li, Shuhan Lv, Fang Lin
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Applied Sciences
Subjects:
Online Access:https://www.mdpi.com/2076-3417/15/7/3862
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850212918497378304
author Yuang Chen
Yong Li
Shaohua Li
Shuhan Lv
Fang Lin
author_facet Yuang Chen
Yong Li
Shaohua Li
Shuhan Lv
Fang Lin
author_sort Yuang Chen
collection DOAJ
description This paper proposes a lightweight violent behavior recognition model, DualCascadeTSF-MobileNetV2, which is improved based on the temporal shift module and its subsequent research. By introducing the Dual Cascade Temporal Shift and Fusion module, the model further enhances the feature correlation ability in the time dimension and solves the problem of information sparsity caused by multiple temporal shifts. Meanwhile, the model incorporates the efficient lightweight structure of MobileNetV2, significantly reducing the number of parameters and computational complexity. Experiments were conducted on three public violent behavior datasets, Crowd Violence, RWF-2000, and Hockey Fights, to verify the performance of the model. The results show that it outperforms other classic models in terms of accuracy, computational speed, and memory size, especially among lightweight models. This research continues and expands on the previous achievements in the fields of TSM and lightweight network design, providing a new solution for real-time violent behavior recognition on edge devices.
format Article
id doaj-art-47e4b1cbed9e431985282d8b7242cd68
institution OA Journals
issn 2076-3417
language English
publishDate 2025-04-01
publisher MDPI AG
record_format Article
series Applied Sciences
spelling doaj-art-47e4b1cbed9e431985282d8b7242cd682025-08-20T02:09:14ZengMDPI AGApplied Sciences2076-34172025-04-01157386210.3390/app15073862DualCascadeTSF-MobileNetV2: A Lightweight Violence Behavior Recognition ModelYuang Chen0Yong Li1Shaohua Li2Shuhan Lv3Fang Lin4Key Laboratory of CTC & IE (Engineering University of PAP), Ministry of Education, Engineering University of PAP, Xi’an 710086, ChinaKey Laboratory of CTC & IE (Engineering University of PAP), Ministry of Education, Engineering University of PAP, Xi’an 710086, ChinaKey Laboratory of CTC & IE (Engineering University of PAP), Ministry of Education, Engineering University of PAP, Xi’an 710086, ChinaKey Laboratory of CTC & IE (Engineering University of PAP), Ministry of Education, Engineering University of PAP, Xi’an 710086, ChinaKey Laboratory of CTC & IE (Engineering University of PAP), Ministry of Education, Engineering University of PAP, Xi’an 710086, ChinaThis paper proposes a lightweight violent behavior recognition model, DualCascadeTSF-MobileNetV2, which is improved based on the temporal shift module and its subsequent research. By introducing the Dual Cascade Temporal Shift and Fusion module, the model further enhances the feature correlation ability in the time dimension and solves the problem of information sparsity caused by multiple temporal shifts. Meanwhile, the model incorporates the efficient lightweight structure of MobileNetV2, significantly reducing the number of parameters and computational complexity. Experiments were conducted on three public violent behavior datasets, Crowd Violence, RWF-2000, and Hockey Fights, to verify the performance of the model. The results show that it outperforms other classic models in terms of accuracy, computational speed, and memory size, especially among lightweight models. This research continues and expands on the previous achievements in the fields of TSM and lightweight network design, providing a new solution for real-time violent behavior recognition on edge devices.https://www.mdpi.com/2076-3417/15/7/3862deep learninglightweight modelsviolence behavior recognitionedge devices
spellingShingle Yuang Chen
Yong Li
Shaohua Li
Shuhan Lv
Fang Lin
DualCascadeTSF-MobileNetV2: A Lightweight Violence Behavior Recognition Model
Applied Sciences
deep learning
lightweight models
violence behavior recognition
edge devices
title DualCascadeTSF-MobileNetV2: A Lightweight Violence Behavior Recognition Model
title_full DualCascadeTSF-MobileNetV2: A Lightweight Violence Behavior Recognition Model
title_fullStr DualCascadeTSF-MobileNetV2: A Lightweight Violence Behavior Recognition Model
title_full_unstemmed DualCascadeTSF-MobileNetV2: A Lightweight Violence Behavior Recognition Model
title_short DualCascadeTSF-MobileNetV2: A Lightweight Violence Behavior Recognition Model
title_sort dualcascadetsf mobilenetv2 a lightweight violence behavior recognition model
topic deep learning
lightweight models
violence behavior recognition
edge devices
url https://www.mdpi.com/2076-3417/15/7/3862
work_keys_str_mv AT yuangchen dualcascadetsfmobilenetv2alightweightviolencebehaviorrecognitionmodel
AT yongli dualcascadetsfmobilenetv2alightweightviolencebehaviorrecognitionmodel
AT shaohuali dualcascadetsfmobilenetv2alightweightviolencebehaviorrecognitionmodel
AT shuhanlv dualcascadetsfmobilenetv2alightweightviolencebehaviorrecognitionmodel
AT fanglin dualcascadetsfmobilenetv2alightweightviolencebehaviorrecognitionmodel