Real-time Mobile Enabled Scheme for Virtual Spectacle Frame Selection

Selection of physical appearance related personal item in online shopping is extremely difficult. This research presents a mobile enabled scheme that provides an advanced online shopping experience of spectacle frames selection. The application is developed to automatically detect the face shape of the user from an image, give suggestions of frames based on the shape, and provide a virtual try-on for the user. Active Shape Model has been used to identify the facial landmark points. Mathematical equations have been designed to analyze and map the landmark points to a face shape. Expert knowledge has been used to recommend frames for each face shape. Haar feature based face and eye detector provided by OpenCV has been used to detect the face and identify the position of eyes for virtual try-on. So far this is the best mobile enabled application for real-time frame selection.