An interesting patent has surfaced today indicating that Microsoft might be planning to enter the mobile dating industry.
The patent, entitled Image-Based Face Search details how Microsoft could be developing complex facial recognition software. The patent explains how a photo could be uploaded from a mobile phone or PC and compared with millions of other pictures to find similar matching faces.
Imagine a dating site that allowed you to upload a photo of a certain hot celebrity and then returned profiles of all the users who looked like them?
Another implementation of the software is a HotOrNot style application – the user is shown a face and then rates it according to how attractive they find the person.
As you can see from the screenshot below the user can say they really like a persons eyes but that they aren’t too keen on the mouth. Microsoft can then take this information and find some matches that are more suited to your tastes. This would have huge potential in the lucrative online dating industry.
From the patent:
A search includes comparing a query image provided by a user to a plurality of stored images of faces stored in a stored image database, and determining a similarity of the query image to the plurality of stored images. One or more resultant images of faces, selected from among the stored images, are displayed to the user based on the determined similarity of the stored images to the query image provided by the user. The resultant images are displayed based at least in part on one or more facial features.
This disclosure is directed to image-based searching, and is shown and described in the context of an image-based face search for an online dating service.
Traditional image retrieval techniques, such as those used in facial-recognition, cannot readily be applied to image-based face retrieval for online dating, since those conventional techniques typically focus on searching for a specific person (i.e., the same person depicted in the image used as the query), not other similar-looking individuals as desired in the online dating arena. Also, traditional image retrieval applications are not concerned with how to evaluate similarities between two persons’ faces, and how to understand what people perceive and evaluate when viewing similar faces.
The image-based search described herein allows users to search profiles using a desired reference or “query” image. This image based search may be implemented as an alterative to the textual queries used on most online dating sites, or may be combined with such textual queries to provide an augmented image-based face retrieval filter.