Recognizing Physical Violence in Visual Media using Deep Learning
| dc.contributor.author | Haleema Muqarrab | |
| dc.date.accessioned | 2025-09-24T11:47:03Z | |
| dc.date.available | 2025-09-24T11:47:03Z | |
| dc.date.issued | 2025 | |
| dc.description.abstract | Recent studies show that there has been a marked increase in the number of violent events occurring in public places like shopping malls, elevators, sports stadiums, and liquor stores. Unfortunately, these instances are mostly discovered after it’s too late. Violence recognition is a critical task in the realm of public safety and security. In this research, we developed a violence recognition system in videos using You Only Look Once (YOLO) based lightweight model YOLOv8n. We compared our proposed approach with different hybrid approaches as well as existing state of the art deep architectures to reach the best approach. The experimental results demonstrate the effectiveness of the proposed method, showing significant improvements in detection accuracy over baseline models. The findings indicate that the proposed approach not only enhances performance, but also offers a scalable and more general solution for real-world applications. The results also signify the supremacy of YOLOv8n for classification tasks over different deep learning approaches. The datasets used were hockey fight dataset and real life violence situations dataset. | |
| dc.identifier.uri | https://escholar.umt.edu.pk/handle/123456789/6853 | |
| dc.language.iso | en | |
| dc.publisher | UMT, Lahore | |
| dc.title | Recognizing Physical Violence in Visual Media using Deep Learning | |
| dc.type | Thesis |
Files
Original bundle
1 - 1 of 1
No Thumbnail Available
- Name:
- Recognizing Physical Violence in Visual Media using Deep Learning.pdf
- Size:
- 5.37 MB
- Format:
- Adobe Portable Document Format
License bundle
1 - 1 of 1
No Thumbnail Available
- Name:
- license.txt
- Size:
- 1.71 KB
- Format:
- Item-specific license agreed upon to submission
- Description: