|10||Assistive Technology for Patients with Medical Face Blindness
|Yuchen He TA||final_paper
|Prosopagnosia is a neurological condition characterized by the inability to recognize faces. After talking a few times with an ECE professor that has this condition, I'd like to work on developing a prototype for a minimally intrusive assist technology designed to help normalize social interactions. The basic idea is to create one piece of wearable tech that captures image data to be sent to a smartphone. The phone can handle facial recognition from a user managed database, and provide the needed information to the user. For example, an ear mounted camera with a subtle activation button might send a photo to the phone, which will identify the largest face in the image and transmit the information to the user through a second piece of wearable tech.
Building on projects FA15_30 and FA12_17, the second wearable is a wristband with a screen, with their WiFi replaced by bluetooth if we keep the phone app, the button I/O for camera activation. Battery and charging for both the camera and wristband prototypes as well. In addition, the wristband buzz once if the face is not in the database, so the user can immediately transition into introductions. A different buzz pattern would indicate that the face was identified, and the information transmitted to the screen.
On the software side, we weren't too interested in duplicating the work already done by so many other labs, and were hoping to just use API calls to any of these http://blog.rapidapi.com/2017/11/10/top-10-facial-recognition-apis-of-2017/ taking advantage of cellular networks and the cloud. This would then free us up to do the more interesting software work of creating the UI to manage the database, and give us time to work on more interesting hardware (like the second piece of wearable tech) which could be more important for this class.