AI Suspicious Activity Detection using Human Pose Estimation

Computer Engineering

Video surveillance plays a central role in today's world. Technologies have become too advanced as artificial intelligence, machine learning, and deep learning invaded the system. Using the above combinations, various systems are in place to help distinguish various suspicious behavior from live tracking images. Human behavior is the most unpredictable and it is very difficult to determine whether it is suspicious or normal. The deep learning approach is used to detect suspicious or normal activity in an academic environment and send an alert message to the appropriate authority when suspicious activity is predicted. Tracking is often done through successive frames extracted from the video. The whole frame is divided into two parts. In the first part, the characteristics are calculated based on the video frames, and in the second part, based on the obtained characteristic classifier, the class is predicted as suspicious or normal.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.