Towards Autonomous Depth Perception for Surveillance in Real World Environments

We describe a system for autonomous depth perception for surveillance in real world environments. With the ever-increasing need for surveillance, human operators are facing a challenging task to track suspicious behavior while looking at a large number of videos in real-time. We present an autonomous suspicious object detection method using depth information obtained from the Microsoft Kinect sensor. The Kinect V2 sensor is capable of tracking six people at any time and provides their skeletal diagrams as output. We use this skeletal joints data to determine the hand position of people. The various hand gestures can be classified by training the system with the help of depth information generated by Kinect sensor. Our algorithm then works to detect and identify objects held in the human hand. The proposed system is compact and the complete video processing can be performed by a low-cost single board computer.