Obviousness

US patent 10776471 describes a facial recognition authentication system that incorporates "path parameters" derived from the movement of a mobile device during image capture. This analysis considers the obviousness of the claimed invention under 35 U.S.C. § 103, based on the general state of the art as described within the patent itself, particularly before the priority date of August 28, 2014.

The patent itself outlines the known limitations of existing facial recognition and biometric systems prior to its invention. Specifically, it notes that "Two-dimensional facial recognition is commonly used to tag people in images on social networks or in photo editing software". However, it explicitly states that "Facial recognition software has not been widely implemented on its own to securely authenticate users attempting to gain access to an account because it not considered secure enough". The primary security concern highlighted is that "two-dimensional facial recognition is considered unsecure because faces may be photographed or recorded, and then the resulting prints or video displays showing images of the user may be used to trick the system". This clearly identifies a "long-felt need" for a more secure facial recognition authentication method.

A person having ordinary skill in the art (PHOSITA) in mobile device security or biometrics, prior to the priority date, would have been familiar with:

• A. 2D Facial Recognition: The technology to detect and recognize faces from images was established, albeit recognized as vulnerable to spoofing when used for security.
• B. Mobile Device Capabilities: Modern mobile devices were commonly equipped with cameras capable of capturing images and video.
• C. Motion Sensors in Mobile Devices: Mobile devices routinely included various movement detecting sensors such as accelerometers, magnetometers, and gyroscopes, used for orientation, gaming, and navigation.
• D. Biometric Authentication Principles: The general concept of enrolling a biometric sample and then comparing subsequent samples for authentication was well-known, as evidenced by the discussion of fingerprint systems.

Motivation for Combination:

The patent itself provides the explicit motivation for combining these elements: to address the critical vulnerability of 2D facial recognition to spoofing attempts using static images or videos. The stated "need for reliable, cost-effective, and convenient method to authenticate users" would compel a PHOSITA to seek solutions that leverage existing, ubiquitous mobile device hardware.

Obvious Combination of Prior Art Elements:

Given the known problems and available technologies, a PHOSITA would have found the following combination obvious:

1. Combining 2D Facial Recognition (A) with Mobile Device Cameras (B) for Liveness Detection: Recognizing the susceptibility of 2D facial recognition to spoofing, a PHOSITA would seek to introduce a "liveness" or 3D aspect to the authentication process. An obvious approach would be to require the user to move the mobile device's camera (B) relative to their face while capturing images. This would generate a sequence of images from different angles and distances, making it significantly harder to spoof with a flat photograph or video. The patent describes this directly: "the user may provide the necessary images (the term images includes video as video is a succession of images) from many different angles and/or positions, and may provide path parameters of the device while obtaining the images ('authentication movement') to both confirm the identity of the user as well as the liveness and realness of that individual to ensure it is not a video, screen shot, or other representation of the person".

2. Integrating Mobile Device Motion Sensors (C) with the Movement Requirement: Once the requirement for device movement during image capture is established, it would be an obvious next step to utilize the existing motion sensors (C) within the mobile device to track and record this movement. The patent states that "the user's device includes a camera and at least one movement detecting sensor, such as an accelerometer, magnetometer, and gyroscope". A PHOSITA would readily recognize that these sensors could provide "path parameters" (such as acceleration, speed, angle, and distance of the mobile device) during the imaging process.

3. Applying Biometric Enrollment and Authentication Principles (D) to Movement Data: In line with standard biometric practices, the recorded movement data ("enrollment movement path") could be stored during an initial enrollment phase. Subsequently, during authentication, a user would be required to perform a similar movement, and the newly recorded "authentication movement" from the sensors would be compared against the stored enrollment movement. This comparison directly addresses the spoofing problem by requiring a dynamic, physical input that cannot be faked by merely presenting a static image or a video of a face without the corresponding device movement. As the patent explains, "if during a login attempt the camera of the mobile device imaged a digital screen displaying a person rotating their head while the phone was not moving, the accelerometer, magnetometer, and gyroscope data would not detect any motion. Thus, the enrollment movement and the authentication movement would not correspond, and the login attempt would be denied".

The combination of readily available mobile device components (camera, accelerometer, gyroscope, magnetometer) with known facial recognition techniques, in an effort to overcome the recognized security deficiencies of 2D facial recognition, would have been obvious to a PHOSITA. The motivation would be to create a more robust and secure authentication system by introducing a dynamic, physical challenge that validates the "liveness" and 3D presence of the user, while leveraging existing hardware to maintain cost-effectiveness and convenience.