Apple files patent for face recognition

The United States Patent and Trademark Office published a patent application by Apple titled Low Threshold Face Recognition. While face recognition is nothing new, the application of the technology on mobile devices is still pretty much a hit and miss affair. Google introduced Face Unlock in Ice Cream Sandwich, but as we have demonstrated, the feature is easily fooled with just a picture.

In its patent application Apple argues:

Most face recognition systems fall into one of two categories. A first category system tends to be robust and can tackle various lighting conditions, orientations, scale and the like, and tends to be computationally expensive. A second category system is specialized for security-type applications and can work under controlled lighting conditions. Adopting the first category systems for face recognition on consumer operated portable appliances that are equipped with a camera would unnecessarily use an appliance’s computing resources and drain its power. Moreover, as the consumer portable appliances tend to be used both indoor and outdoor, the second category systems for face recognition may be ineffective. Such ineffectiveness may be further exacerbated by the proximity of the user to the camera, i.e., small changes in distance to and tilt of the appliance’s camera dramatically distort features, making traditional biometrics used in security-type face recognition ineffective.

Apple attempts to address this problem with a different approach to facial recognition. Instead of the software tracking the entire face, Apple’s Low Threshold Face Recognition analyses certain features of the face. In this case Apple relies on the distance between the eyes and between the eyes and the mouth.

Apple elaborates in its patent application:

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, are disclosed for reducing the impact of lighting conditions and biometric distortions, while providing a low-computation solution for reasonably effective (low threshold) face recognition. In one aspect, the methods include processing a captured image of a face of a user seeking to access a resource by conforming a subset of the captured face image to a reference model. The reference model corresponds to a high information portion of human faces. The methods further include comparing the processed captured image to at least one target profile corresponding to a user associated with the resource, and selectively recognizing the user seeking access to the resource based on a result of said comparing.

The diagram below was submitted by Apple as part of its patent application.

The phrase ‘low threshold’ in the title of the patent means that the method can tolerate a certain level of false positives and still make an accurate face recognition.
While this approach is certainly unique, we don’t see how this can completely eliminate the possibility of a picture compromising the feature. For now, it’s still too early to say if the technology is able to differentiate the difference between an image of a face and the actual face itself.
There are no indications on when Apple will implement this technology in its devices but we won’t be surprised if it makes it to the next big iOS or iPhone/iPad release.

[source]