Suspicious Activity Detection Using CNN
Keywords:
Convolutional Neural Networks (CNNs)Abstract
The security system gets better with
Suspicious Activity Detection using CNNs because this
method automatically detects deviations from normal
behavior through video analysis. A Convolutional
Neural Networks (CNNs) type of deep learning
technology enables staffless video surveillance by
detecting trespassing and loitering incidents along with
aggressive behavior. The security network learns to
identify between commonplace and alarming behaviors
during its operational period. The system builds its
capabilities through video training that combines
regular daily activities with security threats. After the
training process the model demonstrates the capability
to identify security breaches with efficiency. The unique
selling point of this solution is its quick processing time
combined with high accuracy while operating in realtime.
The system functions across diverse environments
because the researchers designed it to operate whether
light conditions change or cameras are situated
differently or background sounds vary. System testing
with genuine surveillance video demonstrated its
operation success throughout multiple realistic
scenarios.
Downloads
References
1. Wang, Limin, Yu Qiao, and Xiaoou Tang. "Video
action detection with relational dynamic-poselets."
In Computer Vision–ECCV 2014: 13th European
Conference, Zurich, Switzerland, September 6-12,
2014, Proceedings, Part V 13, pp. 565-580. Springer
International Publishing, 2014.
2. Hou, Rui, Chen Chen, and Mubarak Shah. "Tube
convolutional neural network (t-cnn) for action
detection in videos." In Proceedings of the IEEE
international conference on computer vision, pp.
5822-5831. 2017.
3. Wang, Lei, Du Q. Huynh, and Piotr Koniusz. "A
comparative review of recent kinect-based action
recognition algorithms." IEEE Transactions on
Image Processing 29 (2019): 15-28.
4. Elahi, GM Mashrur E., and Yee-Hong Yang. "Online
learnable keyframe extraction in videos and its
application with semantic word vector in action
recognition." Pattern Recognition 122 (2022):
108273.
5. Simonyan, K., & Zisserman, A. (2014). "Two-stream
convolutional networks for action recognition in
videos." Advances in Neural Information Processing
Systems (NeurIPS).
6. Carreira, J., & Zisserman, A. (2017). "Quo vadis,
action recognition? A new model and the kinetics
dataset." IEEE Conference on Computer Vision and
Pattern Recognition (CVPR).
7. Tran, D. et al. (2015). "Learning spatiotemporal
features with 3D convolutional networks."
International Conference on Computer Vision
(ICCV).
8. Girdhar, R. et al. (2018). "Detect-and-track: Efficient
pose-based action recognition in videos." IEEE
Conference on Computer Vision and Pattern
Recognition (CVPR).
9. Wang, L. et al. (2016). "Temporal segment networks:
Towards good practices for deep action recognition."
European Conference on Computer Vision (ECCV).
10. Sun, C. et al. (2019). "VideoBERT: A joint model for
video and language representation learning." IEEE
International Conference on Computer Vision
(ICCV).
11. Gavrilyuk, K. et al. (2020). "Actor-transformer: A
transformer architecture for action localization."
British Machine Vision Conference (BMVC).
12. Feichtenhofer, C. et al. (2019). "SlowFast networks
for video recognition." IEEE International
Conference on Computer Vision (ICCV).
13. Arnab, A. et al. (2021). "ViViT: A Video Vision
Transformer." International Conference on Computer
Vision (ICCV).
14. Bertasius, G. et al. (2021). "Space-time attention for
efficient video understanding." International
Conference on Machine Learning (ICML).
