Siam Tahsin Bhuiyan
- Intern, CCDS
Bachelors
Independent University, Bangladesh
Conference Paper2025
Evaluating Pose Estimation’s Contribution to Violence Detection Using Deep Learning
Siam Tahsin Bhuiyan, Mahmudul Haque, Halima Khatun, Rashedur Rahman, Sajed Imtenanul Haque, Ashraful Islam, Saadia Binte Alam
2025 International Conference on Machine Learning and Cybernetics (ICMLC)
In: 2025 International Conference on Machine Learning and Cybernetics (ICMLC)
IEEE, pp. 56-61
Abstract
This study investigates how pose estimation can enhance violence detection using the RWF-2000 dataset and a ConvLSTM-based approach. Pose estimation was integrated into the method by employing MoveNet to extract skeletal key points from video frames, enabling the system to focus on movement patterns for more effective detection of violent actions. Two versions of the model were trained: one using the original dataset and another with a pose-estimated variant. The results demonstrate that the pose-estimation variant achieved higher accuracy, precision, recall, specificity, and F1 score compared to the base version. These improvements indicate that pose estimation enhances the model's ability to interpret movements by minimizing irrelevant background details and emphasizing critical motion patterns. This research highlights the potential of skeletal data to improve the reliability and accuracy of violence detection, supporting advancements in surveillance and action recognition systems.
Keywords
Reliability (semiconductor)PoseDeep learningFocus (optics)Key (lock)Action (physics)Action recognitionMotion (physics)
CCDS Authors
Halima Khatun
- Research Assistant, CCDS
Bachelors
Independent University, Bangladesh
Md. Rashedur Rahman, D. Eng.
- Co-Director, MIRA & NEST Wings
Assistant Professor
Department of Computer Science and Engineering
Independent University, Bangladesh
Ashraful Islam, PhD
- Center Director, CCDS
- Director, HCI Wing
Assistant Professor
Department of Computer Science and Engineering
Independent University, Bangladesh (IUB)Research Interests
Human-Computer Interaction, AI for Social Good, AI for Public Health, AI for Impact
Saadia Binte Alam, PhD
- Director, MIRA & WiSE Wings
Associate Professor
Department of Computer Science and Engineering
Independent University, Bangladesh
References
- 1.Jia Deng, Wei Dong, Richard Socher, Li-Jia Li, Kai Li, Li Fei-Fei. (2009). ImageNet: A large-scale hierarchical image database. 2009 IEEE Conference on Computer Vision and Pattern Recognition, 248–255[10.1109/cvpr.2009.5206848]
- 2.Mark Sandler, Andrew Howard, Menglong Zhu, Andrey Zhmoginov, Liang-Chieh Chen. (2018). MobileNetV2: Inverted Residuals and Linear Bottlenecks. , 4510–4520[10.1109/cvpr.2018.00474]
- 3.Andrej Karpathy, George Toderici, Sanketh Shetty, Thomas Leung, Rahul Sukthankar, Li Fei-Fei. (2014). Large-Scale Video Classification with Convolutional Neural Networks. , 1725–1732[10.1109/cvpr.2014.223]
- 4.Alexander Toshev, Christian Szegedy. (2014). DeepPose: Human Pose Estimation via Deep Neural Networks. , 1653–1660[10.1109/cvpr.2014.214]
- 5.Ronald Poppe. (2009). A survey on vision-based human action recognition. Image and Vision Computing, 28(6), 976–990[10.1016/j.imavis.2009.11.014]
- 6.Yong Du, Wei Wang, Liang Wang. (2015). Hierarchical recurrent neural network for skeleton based action recognition. , 1110–1118[10.1109/cvpr.2015.7298714]
- 7.Ming Cheng, Kunjing Cai, Ming Li. (2021). RWF-2000: An Open Large Scale Video Database for Violence Detection. , 4183–4190[10.1109/icpr48806.2021.9412502]
- 8.Muhammad Ramzan, Adnan Abid, Hikmat Ullah Khan, Shahid Mahmood Awan, Amina Ismail, Muzamil Ahmed, Mahwish Ilyas, Ahsan Mahmood. (2019). A Review on State-of-the-Art Violence Detection Techniques. IEEE Access, 7, 107560–107575[10.1109/access.2019.2932114]
- 9.Soheil Vosta, Kin-Choong Yow. (2022). A CNN-RNN Combined Structure for Real-World Violence Detection in Surveillance Cameras. Applied Sciences, 12(3), 1021[10.3390/app12031021]
- 10.Shuhua Liu, Thomas Forss. (2015). New Classification Models for Detecting Hate and Violence Web Content. , 487–495[10.5220/0005636704870495]
- 11.Chaonan Gu, Xiaoyu Wu, Shengjin Wang. (2020). Violent Video Detection Based on Semantic Correspondence. IEEE Access, 8, 85958–85967[10.1109/access.2020.2992617]