Skip to main content

Research

Project Title: 2D-3D Face Recognition

Effectively screening and identifying individuals against both known and unknown perpetrators is encumbered by the unconstrained pose, arbitrary illumination conditions, and resolution mismatches present in the image and video data compared. This problem is compounded when the image or video of an individual is acquired at a distance using trail cameras, which is often the case at US borders. To address these challenges, our approach will develop new methods and integrate existing solutions as needed to improve overall operational and performance capabilities. Our goal is to develop accurate, robust and efficient 3D-aided face recognition algorithms from image and/or video for identification in adverse outdoor conditions.

2d-3d-face-recognition.png

 

Project Title: iRay

Project Duration: 2015 - current

Description: iRay project aims to design and develop a prototype for a mobile augmented reality medical application in purpose of superimposing the pre-operative 3D anatomical data on patient’s body in a real-time video. The proposed mobile system applies depth information to register the virtual anatomical data to the real patient’s surface. It can help physicians intuitively observe the inner organs and make better decisions before or during the interventions.

Demos:

Demo1: Prototype system on iPad

Demo2: Accuracy measurement on iPad

Demo3: Exploration on HoloLens