NYUAD Junior Explores Future of Surveillance During CSE Internship

A visual representation of the 3D depth data from a snapshot of readings at a particular moment in time.

At the end of NYU Abu Dhabi's spring 2013 semester, after the most academically and personally demanding term of my undergraduate career came to a close, I had but two days to relax before my summer engineering classes began. The courses — Computer Systems Programming and Advanced Digital Logic — engaged me like no others. I was interested in the things I was being taught and enjoyed exploring the topics. I wanted to know the boundaries of the concepts.

It was in this state of mind that I started my summer internship, working on a research project at the University's Center for Science and Engineering (CSE) under NYUAD Assistant Professor of Electrical Engineering Mohamad Eid.

The aim of the project was to use 3D depth coordinates from an infrared sensor to track the joints of a human and generate a skeleton, and then to use the variations in the generated skeletons to monitor human activities. Through classifying certain movements and activities as "suspicious" or "normal" we aimed to devise a system that could be used for remote surveillance purposes. Examples of suspicious activities could range from entering a room outside of office hours, standing on a table, or falling down. Should a suspicious activity occur, a stream to an external server would be triggered so that security personnel could monitor the individual using depth plots, standard RGB images, and sound.

It was exciting to put theories and lessons from my classes into action in a real-life project.

Umair Saad, Class of 2015

The promise of such a system is significant. Contemporary surveillance systems use closed-circuit cameras that require continuous monitoring, but tracking using depth data that sounds an alarm to denote suspicious activity would enable security personnel to focus on a particular screen at the right times, and may perhaps one day lead to autonomous detection of unauthorized behavior in a given space.

At the beginning of the project I didn't know where to start; however, the research assistants were a great help in teaching me how to search for and begin to understand things I didn't know. It was exciting to put theories and lessons from my classes into action in a real-life project. Some of them were incredibly useful and the experience put class teachings into better perspective.

After successfully completing the project and setting it up at the CSE, I flew home to Islamabad for the remainder of the summer. While there, I learned that a paper I wrote with classmate Usama Afzal — "A Real Time Vibrotactile Biofeedback System for Optimizing Athlete Training" — was accepted for publication and presentation at the 2013 IEEE HAVE (Haptic Audio-Visual Environments and Games) conference in Istanbul at the end of October.

I'm now studying in New York for the fall semester while working on another research project, looking forward to tackling the Capstone project later this year and to continue working on projects that I feel contribute to the real world. With these goals in mind, I'm ready to take on this city.