Two-Stage Video Violence Detection Framework Using GMFlow and CBAM-Enhanced ResNet3D

Mahmoud, Mohamed; Yagoub, Bilel; Senussi, Mostafa Farouk; Abdalla, Mahmoud; Kasem, Mahmoud Salaheldin; Kang, Hyun-Soo

doi:10.3390/math13081226

Open AccessArticle

Two-Stage Video Violence Detection Framework Using GMFlow and CBAM-Enhanced ResNet3D

by

Mohamed Mahmoud

^1,2

,

Bilel Yagoub

¹,

Mostafa Farouk Senussi

^1,2

,

Mahmoud Abdalla

¹

,

Mahmoud Salaheldin Kasem

^1,3

and

Hyun-Soo Kang

^1,*

¹

Department of Information and Communication Engineering, School of Electrical and Computer Engineering, Chungbuk National University, Cheongju-si 28644, Republic of Korea

²

Information Technology Department, Faculty of Computers and Information, Assiut University, Assiut 71526, Egypt

³

Multimedia Department, Faculty of Computers and Information, Assiut University, Assiut 71526, Egypt

^*

Author to whom correspondence should be addressed.

Mathematics 2025, 13(8), 1226; https://doi.org/10.3390/math13081226

Submission received: 20 January 2025 / Revised: 15 March 2025 / Accepted: 7 April 2025 / Published: 8 April 2025

(This article belongs to the Special Issue Advanced Research in Neural Networks, Machine Learning, and Image Processing)

Download

Browse Figures

Versions Notes

Abstract

Video violence detection has gained significant attention in recent years due to its applications in surveillance and security. This paper proposes a two-stage framework for detecting violent actions in video sequences. The first stage leverages GMFlow, a pre-trained optical flow network, to capture the temporal motion between consecutive frames, effectively encoding motion dynamics. In the second stage, we integrate these optical flow images with RGB frames and feed them into a CBAM-enhanced ResNet3D network to capture complementary spatiotemporal features. The attention mechanism provided by CBAM enables the network to focus on the most relevant regions in the frames, improving the detection of violent actions. We evaluate the proposed framework on three widely used datasets: Hockey Fight, Crowd Violence, and UBI-Fight. Our experimental results demonstrate superior performance compared to several state-of-the-art methods, achieving AUC scores of 0.963 on UBI-Fight and accuracies of 97.5% and 94.0% on Hockey Fight and Crowd Violence, respectively. The proposed approach effectively combines GMFlow-generated optical flow with deep 3D convolutional networks, providing robust and efficient detection of violence in videos.

Keywords: video violence detection; GMFlow; optical flow; CBAM (convolutional block attention module); ResNet3D; anomaly detection

Share and Cite

MDPI and ACS Style

Mahmoud, M.; Yagoub, B.; Senussi, M.F.; Abdalla, M.; Kasem, M.S.; Kang, H.-S. Two-Stage Video Violence Detection Framework Using GMFlow and CBAM-Enhanced ResNet3D. Mathematics 2025, 13, 1226. https://doi.org/10.3390/math13081226

AMA Style

Mahmoud M, Yagoub B, Senussi MF, Abdalla M, Kasem MS, Kang H-S. Two-Stage Video Violence Detection Framework Using GMFlow and CBAM-Enhanced ResNet3D. Mathematics. 2025; 13(8):1226. https://doi.org/10.3390/math13081226

Chicago/Turabian Style

Mahmoud, Mohamed, Bilel Yagoub, Mostafa Farouk Senussi, Mahmoud Abdalla, Mahmoud Salaheldin Kasem, and Hyun-Soo Kang. 2025. "Two-Stage Video Violence Detection Framework Using GMFlow and CBAM-Enhanced ResNet3D" Mathematics 13, no. 8: 1226. https://doi.org/10.3390/math13081226

APA Style

Mahmoud, M., Yagoub, B., Senussi, M. F., Abdalla, M., Kasem, M. S., & Kang, H.-S. (2025). Two-Stage Video Violence Detection Framework Using GMFlow and CBAM-Enhanced ResNet3D. Mathematics, 13(8), 1226. https://doi.org/10.3390/math13081226

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu

Two-Stage Video Violence Detection Framework Using GMFlow and CBAM-Enhanced ResNet3D

Abstract

Share and Cite

Article Metrics

Article Access Statistics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI