The NovelHAD algorithm to detect presence of human activity in videos based on MediaPipe pose and human landmarks
Video surveillance is one of the best tools available today for detecting and preventing workplace fraud and other unwanted activity. This study aims to find the presence of human activity (no activity or activity) in the CCTV video footage. This study develops a novel algorithm called NovelHAD base...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2025-12-01
|
| Series: | Cogent Social Sciences |
| Subjects: | |
| Online Access: | https://www.tandfonline.com/doi/10.1080/23311886.2025.2474863 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Video surveillance is one of the best tools available today for detecting and preventing workplace fraud and other unwanted activity. This study aims to find the presence of human activity (no activity or activity) in the CCTV video footage. This study develops a novel algorithm called NovelHAD based on the human kinematic prototype anatomy and the MediaPipe Pose (MPP) landmark detector feature extraction technique. The movements toward a person’s hands and legs indicate human activity. The distances between the human landmarks or feature vectors (LEFT_HIP, LEFT_ELBOW), (RIGHT_HIP, RIGHT_ELBOW), and (LEFT_ANKLE, RIGHT_ANKLE) were measured in Euclidean space to capture the motions. The UCF-101 Crime public dataset is used to investigate the presence of human activity. The experiment’s findings indicate a 95% accuracy. |
|---|---|
| ISSN: | 2331-1886 |