A Holistic Approach to Robust and Scalable Object Recognition

This project is conducted at Advanced Digital Sciences Center (ADSC), as part of the Interactive Digital Media (IDM) sub-program. The PI leading our research efforts is Dr. Yi Ma from UIUC and MSRA.

imaging_system

[Left plot] Training acquisition system: Four projectors and two cameras controlled by one computer. [Top-right plot] 38 traning images of one subject captured by the training acquisition system. [Bottom-right plot] Representative testing examples, including indoor/outdoor images (1st row), images of subjects with eyeglasses (2nd row), and image of subjects with sunglasses (3rd row).

 

ADSC's interative digital media (IDM) seeks to incorporate audio, video, and unstructured web data to provide a genuine human sixth sense in augmented reality and related applications. Object/face recognition is a key capability for supporting this sixth sense. For example, an augmented reality system must recognize each object that it will provide additional information about. As another example, reliable face recognition can enable personalization of the environment in a smart office or home, from customized heating and cooling settings to automated toddler locks for kitchen appliances and reminders of upcoming appointments at the office.  ADSC's integrative applications for virtual meetings and smart offices leverage this ability to recognize people and other objects.  For example, when a virtual meeting system recognizes a particular person at a meeting, the entity search facilities of the ARISE project can find, disambiguate, reconcile, and summarize key information about that person, and make that information available to other attendees.