June 30, 2026
On June 4, 2026, the PTAB issued a final written decision in a petition (IPR2025-00109, Jumio Corporation) filed against Patent 11,874,910 B2 to FaceTec, Inc., determining, in a majority opinion, that no challenged claims of claims 1-24 are unpatentable under 35 U.S.C. § 318(a).
The FaceTec, Inc. patent has three independent claims, claims 1, 10 and 22. Independent claim 1, representative of the limitations under challenge, is presented below:
1. A computing device for verifying three-dimensionality of a user via a user's camera equipped computing device, the computing device comprising:
a processor configured to execute machine executable code;
a screen configured to provide a user interface to the user;
a camera configured to capture images;
one or more memories configured to store machine readable instructions that are stored on the memory of the authentication server which when executed by the processor, cause the computing device to:
capturing at least one first image of the user taken with the camera of the computing device at a first location which is a first distance from the user;
processing the at least one first image or a portion thereof to create first data;
capturing at least one second image of the user taken with the camera of the computing device is at a second distance from the user, the second distance being different than the first distance, the capturing at least one second image of the user occurring after movement of the camera or the user to establish the camera at the second distance from the user;
processing the at least one second image or a portion thereof to create second data;
comparing the first data to the second data to determine whether expected differences exist between the first data and the second data which indicates three-dimensionality of the user;
verifying the images of the user exhibit three dimensional traits when the expected differences exist between the first data and the second data as a result of capturing the at least one first image and the at least one second image at different distances from the user.
The FaceTec patent describes an authentication method based on facial recognition using images sourced at different distances from the user in order to determine three dimensionality of the image. The images are transmitted to an authentication center to determine that the face in the image is three-dimensional, i.e. not a representation on a printed picture or a video screen. The authentication biometrics are then compared to enrollment biometrics to verify the user enrollment.
The Decision states (Pages 6-7) that the FaceTec patent explains that two-dimensional facial recognition is unsecure because faces may be photographed or recorded and resulting prints or video displays of a user may be used to spoof the system. The FaceTec patent observes which images are taken at relatively close distances between a subject and a camera (such as a mobile phone camera), where the images are distorted by “perspective distortion”. Embodiments take advantage of this distortion effect to increase security of an authentication system. In such embodiments, images are obtained at both close and far distances during authentication. Because perspective distortion is expected when an actual, three-dimensional person is present, an absence of the relative change in the dimensions of the facial features alerts the system to a fraudulent attempt at authentication. According to the FaceTec patent, the distortion effect could not easily be re-created with a two-dimensional picture and thus embodiments provide a secure test to prevent a two-dimensional picture (in place of a 3D face) from being used for authentication.
The Petitioner, Jumio, asserted that:
1. Independent claims 1, 10 and 22 are obvious under 35 U.S.C. § 103 based on the prior art of Derakhshani et al., US 8,437,513 B1, issued May 7, 2013. Ex. 1005 (“Derakhshani”) and Tanii et al., US 2002/0113884 A1, published Aug. 22, 2002. Ex. 1007 (“Tanii”), and
2. Independent claims 1, 10 and 22 are obvious under 35 U.S.C. § 103 based on the prior art of Zhang et al., US 2011/0299741 A1, published Dec. 8, 2011. Ex. 1006 (“Zhang”) and Tanii.
1. Combination of Derakhshani and Tanii (Decision, Page 14)
Derakhshani describes embodiments that “may prevent spoofing of an eye biometric based authentication system using objects that are not a living human eye.” Derakhshani describes verifying whether a user’s face is 3D instead of a 2D photo by obtaining a plurality of images at different camera focus settings to determine distances to various facial landmarks and determining whether such distances deviate from a 2D plane. Derakhshani describes an alternative embodiment where a camera obtains a plurality of images from different perspectives and the images are examined to determine whether the images exhibit parallax effects indicative of 3D objects.
Petitioner acknowledges that Derakhshani does not describe a system that verifies that captured user images are three-dimensional based on perspective distortion and, so, turns to Tanii.
Tanii relates to digital photography and image processing, including “technology to correct warp of a captured image.” Tanii explains that Figure 4B shows a warped, distorted, or “unnatural image [that] results in which the perspective is exaggerated” because the user and the cell phone capturing the image of the user “are close together.” Tanii explains that such distortions can be corrected by “enlarging” the peripheral areas of the image relative to the center area.·····················································
Tanii is used for the fact that although Derakhshani teaches the focus distance approach and the parallax approach, which two approaches can be used for determining three dimensionality, Tanii teaches a third approach, which is distance-induced distortions.
The Decision (Pages 16-19) finds that the combination of Derakhshani and Tanii has not provided adequate motivation to support its proffered combination and modification of Derakhshani with Tanii.
Tanii, for its part, does not describe a technique for verifying that images are three-dimensional—let alone a technique that captures two images at different distances, compares the images for certain differences, and verifies three dimensionality based on those differences, as recited by the claim. Instead, Tanii describes a system that “corrects warp in which the perspective is exaggerated when an image is captured with the user’s face at close range.
Petitioner offers a number of reasons to combine Derakhshani and Tanii: “Tanii merely teaches an obvious alternative to Derakhshani’s existing two approaches to evaluate whether a face being captured is three-dimensional”. Petitioner also asserts there would have been motivation to combine because “evaluating for distance-induced distortion consistent with Tanii would be easier for users on a mobile device” or useful for mobile device with “wide angle lenses.”
The Decision finds that these reasons would not have adequately motivated a person having ordinary skill in the art to combine and modify the references in Petitioner’s asserted manner to reach the claimed features. Petitioner’s motivations -“substitution,” “obvious alternative . . . approaches,” “use of a known technique,” and “complementary” techniques - are all similarly based on swapping or adding techniques for verifying three dimensionality. See Pet. 13–17; Pet. Reply 16–17. Those motivations might have been suitable had Tanii described techniques to verify three dimensionality, but Petitioner has not shown that Tanii describes any technique to verify three dimensionality. See Pet. 13–17, 25–28. Instead, as discussed above, Tanii describes the existence of the perspective distortion phenomenon and a technique for correcting it. As such, substituting the knowledge that perspective distortion exists and techniques to correct it does not correspond to a technique for correcting it, and so does not adequately motivate Petitioner’s combination. Put another way, because Tanii does not teach an approach for verifying three-dimensionality, Tanii’s techniques cannot be used as a substitute for approaches that do verify three dimensionality simply because Tanii discloses the concept of perspective distortion.
The Decision states (Page 20): Additionally, Petitioner’s motivations do not support the multiple modifications required to bridge the gap between recognizing the phenomenon of perspective distortion and applying that phenomenon to verify three dimensionality via the claimed technique.
The Decision states (Page 22): Accordingly, for the reasons discussed above, we determine that Petitioner fails to show by a preponderance of the evidence that a person of ordinary skill in the art would have been motivated to combine and modify Derakhshani and Tanii as proposed. Every challenged claim over the combination of Derakhshani and Tanii relies on the same flawed motivation and, as such, we determine that Petitioner fails to show by a preponderance of the evidence that claims 1–24 would have been obvious over Derakhshani and Tanii or Derakhshani, Tanii and Tahk.
2. Combination of Zhang and Tanii
Zhang (Decision, Pages 11-12) generally relates to “distinguishing live faces from flat surfaces.” Zhang describes a 3D structure determination module 112 that uses a homography based technique for determining whether a face is a 3D structure or a flat surface. “The homography based technique leverages the fact that two views of a flat (planar) surface are related based on a homography matrix H.”
Petitioner (Pages 22-24) asserts that a person of ordinary skill in the art would have been motivated to combine Zhang and Tanii to: “enhance Zhang’s process”, “take advantage of the distance-induced distortions identified by Tanii”, “reduce spoofing of Zhang’s system”, “provide a simpler mathematical operation that may require less computational resources”, “take advantage of these distance induced distortions to distinguish between three-dimensional faces and two dimensional pictures” and “to utilize these expected distortions as either a supplemental or alternative verification of three-dimensionality of a face”.
For similar reasons as our discussion of Derakhshani and Tanii above, Petitioner’s proffered motivations do not adequately support its combination of Zhang and Tanii. As discussed above, Petitioner’s stated motivations rely on an insight that Tanii does not teach, namely, using perspective distortion to identify or verify three dimensionality. As such, those motivations do not match Petitioner’s combination and modifications.
Furthermore, Petitioner’s combination requires modifications that its proffered motivations do not support. In one suggestion, Petitioner asserts its combination “perform[s] Zhang’s three-dimensional verification process, but rather than ‘warp’ one image using a homography transform to compare it to the second image and evaluate the two for differences, a POSITA would ‘warp’ (i.e., correct) the ‘close’ image using Tanii’s distortion-correction procedure to compare it to the ‘far’ image and evaluate the two for differences.” In another suggestion, Petitioner asserts “a POSITA would have been motivated to use Zhang’s existing process, but take advantage of the distance-induced distortions identified by Tanii by using a (wide-angle) camera that induces these distortions at different distances, and capturing two images of the face at different distances.” But, as Patent Owner and a Dr. Turk (Ex. 2061) point out, there are unacknowledged complexities to those modifications which seemingly “create features out of thin air during distortion correction in an attempt to compare it to a second photo,” e.g., ears missing from one image which are used in another image. Similarly, Patent Owner and Dr. Turk point out that Petitioner’s modifications must compare images based on perspective distortions and, so, require “manufacturing an entire set of operations and evaluations that do not appear in either reference.” As above, Petitioner’s motivations do not support its leap from the knowledge of the perspective distortion phenomenon to the use of that knowledge to verify three dimensionality. Further, such added complexity suggests that Petitioner’s proffered motivations inadequately support the combination.
Accordingly, for the reasons discussed above, we (the PTAB) determine that Petitioner fails to show by a preponderance of the evidence that a person of ordinary skill in the art would have been motivated to combine and modify Zhang and Tanii as proposed. Every challenged claim over the combination of Zhang and Tanii relies on the same flawed motivation and, as such, we determine that Petitioner fails to show by a preponderance of the evidence that claims 1–24 would have been obvious over Zhang and Tanii.
Decision, (Page 29): Independent claims 10 and 22 include limitations similar to limitation. We (the PTAB), therefore, determine that the prior art does not teach these limitations of all independent claims of the ’910 Patent, and for the same reasons, all dependent claims of the ’910 Patent.
As discussed above, Petitioner has not set forth adequate motivation to support its challenges based on the combination of Derakhshani and Tanii or Zhang and Tanii nor has Petitioner shown that limitation 1[d3] is taught by any of the references. Although these failures alone defeat Petitioner’s challenges, Patent Owner’s objective indicia, i.e., secondary considerations, are also probative of non-obviousness. In particular, Patent Owner’s evidence on commercial success, industry praise, and skepticism weigh heavily in favor of non-obviousness.
Decision, (Page 34): Based on the evidence presented with the Petition, the evidence introduced during the trial, and the parties’ respective arguments, Petitioner has not shown by a preponderance of the evidence that any of claims 1–24 are unpatentable.
Decision, (Page 36): It is ORDERED that claims 1–24 of U.S. Patent No. 11,874,910 are not determined to be unpatentable.