DCT
3:24-cv-03623
FaceTec Inc v. Jumio Corp
I. Executive Summary and Procedural Information
- Parties & Counsel:
- Plaintiff: FaceTec, Inc. (Delaware)
- Defendant: Jumio Corporation (Delaware)
- Plaintiff’s Counsel: One LLP
- Case Identification: 3:24-cv-03623, N.D. Cal., 06/14/2024
- Venue Allegations: Venue is alleged to be proper based on Defendant Jumio Corporation having a regular and established place of business in the Northern District of California, specifically its headquarters in Sunnyvale.
- Core Dispute: Plaintiff alleges that Defendant’s biometric identity verification and liveness detection software infringes four patents related to authenticating a user by analyzing images of their face captured at different distances.
- Technical Context: The technology concerns software-based "liveness detection" used to prevent spoofing in biometric authentication systems, a critical security feature for online banking, payments, and identity verification.
- Key Procedural History: The complaint alleges that Defendant Jumio previously licensed Plaintiff FaceTec's patented technology before terminating the contract and later contracting with a competitor, iProov Ltd. FaceTec states it is currently suing iProov in the District of Nevada for infringement of two of the same patents asserted in this case. The complaint also notes that FaceTec sent a cease-and-desist letter to Jumio in August 2023, which Jumio allegedly refused.
Case Timeline
| Date | Event |
|---|---|
| 2014-06-10 | '471 Patent Issued (original issue date) |
| 2014-08-28 | '471 Patent Priority Date |
| 2014-08-28 | '938 Patent Priority Date |
| 2017-02-17 | '606 Patent Priority Date |
| 2020-03-12 | '910 Patent Priority Date |
| 2020-09-15 | '471 Patent Issued (re-issue date) |
| 2021-04-07 | Jumio announces partnership with iProov |
| 2021-10-26 | '606 Patent Issued |
| 2021-12-01 | FaceTec sues iProov in D. Nev. (approx. date) |
| 2023-07-04 | '938 Patent Issued |
| 2023-08-23 | FaceTec sends cease-and-desist letter to Jumio |
| 2024-01-16 | '910 Patent Issued |
| 2024-06-14 | Complaint Filing Date |
II. Technology and Patent(s)-in-Suit Analysis
U.S. Patent No. 10,776,471 - “Facial Recognition Authentication System Including Path Parameters”
The Invention Explained
- Problem Addressed: The patent’s background section describes the problem that conventional two-dimensional facial recognition is insecure because it can be tricked by photographs or video displays of a user’s face (’471 Patent, col. 1:62-67). It also notes the inconvenience of typing long, complex passwords on small electronic devices (Compl. ¶8; ’471 Patent, col. 1:39-44).
- The Patented Solution: The invention claims to solve this problem by authenticating a user not just by what their face looks like, but by confirming they are a live, three-dimensional person. It does this by capturing images of the user's face at two different distances (e.g., far and close-up) and analyzing the differences between the images (’471 Patent, col. 2:10-26). The natural optical distortion from a camera lens (a "fish-eye" effect) causes a 3D object to appear differently at close range versus far range; the system verifies that this expected distortion is present, which would not be the case for a 2D photograph (’471 Patent, col. 4:51-56). This process allows the system to validate the user's "liveness" and three-dimensionality. The patent also describes using movement sensor data (e.g., from an accelerometer or gyroscope) to record the "path parameters" of the device as it moves, adding another layer of security (’471 Patent, col. 1:12-14).
- Technical Importance: The approach allows for 3D liveness detection using standard 2D cameras found on nearly all smartphones and computers, avoiding the need for specialized hardware like infrared or depth-sensing cameras (Compl. ¶17).
Key Claims at a Glance
- The complaint asserts at least Claim 10, which depends on independent Claim 1 (Compl. ¶35).
- The essential elements of independent Claim 1 (a system claim) are:
- Capturing a first image of a user at a first distance.
- Processing the first image to create first data.
- Moving the camera to a second location at a different distance from the user.
- Capturing a second image of the user at the second distance.
- Processing the second image to create second data.
- Comparing the first and second data to determine if "expected differences exist" that indicate three-dimensionality.
- Authenticating the user when the data shows "expected distortion resulting from movement."
- The complaint reserves the right to assert other claims (Compl. ¶35).
U.S. Patent No. 11,157,606 - “Facial Recognition Authentication System Including Path Parameters”
The Invention Explained
- Problem Addressed: The ’606 Patent addresses the same technical problem as the ’471 Patent: the insecurity of 2D facial recognition systems against spoofing attacks using photographs or videos (’606 Patent, col. 1:65-col. 2:2).
- The Patented Solution: The patented method is a refinement of the same core concept: using images captured at different distances to confirm 3D liveness. This patent focuses on a method where biometric data from a close-up image and a far-away image are compared. The method claims that for a real 3D person, the biometric data will not match due to optical distortion, whereas for a 2D photograph, the data would match (’606 Patent, col. 4:51-58). The method authenticates the user specifically when the biometric data from the two images do not match but still exhibit the "expected differences" consistent with a 3D object being moved relative to a camera lens (’606 Patent, Claim 1).
- Technical Importance: This patent claims a specific logical method for implementing 3D liveness detection, focusing on the non-matching nature of biometric data from different distances as an affirmative indicator of a live user (Compl. ¶13-14).
Key Claims at a Glance
- The complaint asserts at least independent Claim 1 (Compl. ¶48).
- The essential elements of independent Claim 1 (a method claim) are:
- Capturing a first image of a user's face at a first distance.
- Processing it to obtain first biometric data.
- Capturing a second image of the user's face at a different, second distance.
- Processing it to obtain second biometric data.
- Comparing the first and second biometric data.
- Determining the user's face is three-dimensional specifically when (1) the first biometric data does not match the second, and (2) the second data has "expected differences" compared to the first resulting from the change in distance.
- The complaint reserves the right to assert other claims (Compl. ¶48).
U.S. Patent No. 11,693,938 - “Facial Recognition Authentication System Including Path Parameters”
- Technology Synopsis: This patent covers a method for determining a user does not exhibit three-dimensionality. It teaches capturing images at different distances and processing them to create data. The method determines a lack of 3D characteristics if the data from the different images does not show an "expected type of distorting change" consistent with a real person being imaged (’938 Patent, Abstract; Claim 8).
- Asserted Claims: At least independent Claim 8 is asserted (Compl. ¶61).
- Accused Features: The complaint alleges that Jumio's products, which capture and compare face images from multiple distances, infringe this patent (Compl. ¶¶28d, 61).
U.S. Patent No. 11,874,910 - “Facial Recognition Authentication System Including Path Parameters”
- Technology Synopsis: This patent claims a system and method for verifying three-dimensionality by capturing first and second images at different distances and comparing the resulting biometric data. Similar to the ’606 Patent, it relies on the principle that biometric data from the two images will not match for a live person but will have expected differences due to optical distortion, thereby confirming liveness (’910 Patent, Abstract; Claim 1).
- Asserted Claims: At least independent Claim 10 is asserted (Compl. ¶74).
- Accused Features: The complaint accuses Jumio's liveness detection technology, which prompts users to position their face at multiple distances for image capture and comparison, of infringement (Compl. ¶¶28d, 74).
III. The Accused Instrumentality
Product Identification
The complaint identifies the accused products collectively as the "Jumio Accused Instrumentalities" or "Jumio Liveness Detection Technology" (Compl. ¶28). This includes all Jumio products and services that incorporate technology from iProov Ltd. (e.g., "Liveness Assurance") as well as Jumio's own-branded technologies referred to as "Liveness Checks" or "Liveness Detection" (Compl. ¶¶28a, 28b, 28c).
Functionality and Market Context
The accused functionality involves a user authentication process that: (1) prompts a user to position their face at more than one distance from a device's camera; (2) captures face image data at two or more of these distances; and (3) compares the resulting image data to evaluate the user's three-dimensionality for liveness detection (Compl. ¶28d). The complaint provides visual evidence simulating this process, where a "First Image" is captured from a distance and a "Second Image" is captured closer up, showing the perspective distortion that the technology allegedly analyzes (Compl. p. 6). FaceTec alleges that Jumio is a direct competitor and that its liveness detection products compete with FaceTec's own offerings (Compl. ¶30).
IV. Analysis of Infringement Allegations
U.S. Patent No. 10,776,471 Infringement Allegations
| Claim Element (from Independent Claim 1) | Alleged Infringing Functionality | Complaint Citation | Patent Citation |
|---|---|---|---|
| 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; | Jumio's software prompts a user to position their face and captures a facial image at a first distance from the device camera. | ¶28d | col. 12:28-33 |
| moving the camera from the first location to a second location...to change the distance between the user and the camera...; | Jumio's software prompts the user to move the device (or their head) to a second position at a different distance from the camera. | ¶28d | col. 12:45-50 |
| capturing at least one second image of the user taken with the camera...at the second distance..., the second distance being different than the first distance; | Jumio's software captures a second facial image at the new, different distance from the device camera. | ¶28d | col. 12:28-33 |
| comparing the first data to the second data to determine whether expected differences exist...which indicated three-dimensionality of the user; | Jumio's software compares the face image data from the two captures to evaluate user 3-dimensionality based on the presence of perspective distortion. | ¶¶13, 28d | col. 14:55-64 |
| authenticating the user when differences between the first data and the second data have expected distortion resulting from movement... | Jumio's system uses the confirmation of 3-dimensionality to verify the physical presence of the user as part of an authentication process. | ¶¶28d, 40 | col. 16:26-34 |
Identified Points of Contention
- Scope Questions: A central question may be the construction of "path parameters." The patent describes this term as including data from movement sensors like accelerometers (’471 Patent, col. 6:39-46). The complaint's allegations focus on comparing images from different distances (Compl. ¶28d). The case may turn on whether the accused system must also capture and compare "path parameters" from motion sensors to infringe, or if the change in distance between two images is itself sufficient to meet the claim limitations.
- Technical Questions: What evidence does the complaint provide that Jumio's software performs the specific comparison recited in the claims to find "expected distortion"? The complaint alleges this function (Compl. ¶28d), but a key factual question for the court will be how the accused system's comparison algorithm actually operates.
U.S. Patent No. 11,157,606 Infringement Allegations
| Claim Element (from Independent Claim 1) | Alleged Infringing Functionality | Complaint Citation | Patent Citation |
|---|---|---|---|
| capturing at least one first image of the user's face...at a first distance from the user; | Jumio's software prompts a user to position their face and captures a facial image at a first distance from the device camera. | ¶28d | col. 13:25-31 |
| capturing at least one second image of the user's face...at a second distance from the user, the second distance being different from the first distance; | Jumio's software prompts the user to change the distance to the camera and captures a second facial image. | ¶28d | col. 13:25-31 |
| processing the...first image to obtain first biometric data...processing the...second image to obtain second biometric data; | Jumio's system processes the captured images to extract data for comparison. | ¶28d | col. 13:32-39 |
| comparing the first biometric data with the second biometric data...; | Jumio's software compares the data derived from the first and second images. | ¶28d | col. 14:26-28 |
| determining that the user's face is three-dimensional when: the first biometric data does not match the second biometric data; and the second biometric data has expected differences as compared to the first biometric data... | Jumio's software allegedly evaluates 3-dimensionality by confirming that the biometric data from the two different distances do not match but exhibit predictable distortion, thus verifying a live user. | ¶¶13, 28d | col. 4:51-58 |
Identified Points of Contention
- Scope Questions: The claim requires a specific two-part logical test: (1) the biometric data must not match, AND (2) the data must have "expected differences." A dispute may arise over whether Jumio's system performs this exact two-part determination or uses a different logic (e.g., a simple correlation score) to assess liveness.
- Technical Questions: What constitutes "biometric data" in the accused system, and how does it define a "match"? The patent defines biometric data broadly to include facial features (’606 Patent, col. 6:35-39). The infringement analysis will depend on the specific type of data Jumio's system extracts and the threshold it uses to determine whether two sets of data match or not.
V. Key Claim Terms for Construction
The Term: "expected differences...which indicated three-dimensionality" (’471 Patent, Claim 1) / "expected differences as compared to the first biometric data" (’606 Patent, Claim 1)
- Context and Importance: This term is the core of the invention. Its definition will determine what kind of comparison the accused product must perform to infringe. Practitioners may focus on this term because the plaintiff will likely argue it broadly covers any analysis of perspective distortion, while the defendant may argue it requires a specific, quantitative comparison against a pre-defined model of optical distortion.
- Intrinsic Evidence for Interpretation:
- Evidence for a Broader Interpretation: The specification states that when a 3D person images themselves, "the biometric results are different due to the fish-eye effect of the lens" (’471 Patent, col. 4:51-54). This language could support a broad interpretation where any detected difference attributable to optical effects qualifies as an "expected difference."
- Evidence for a Narrower Interpretation: The detailed description discusses analyzing "incrementally different biometric results" and tracking the "morphing of the face" from far to close-up (’471 Patent, col. 4:60-64). This could support a narrower reading requiring the system to track a specific, progressive distortion pattern rather than just comparing two static endpoints.
The Term: "path parameters" (’471 Patent, Title, Abstract, Claims)
- Context and Importance: Although not in the asserted independent claim, this term is central to the patent's disclosure and title. Its relevance to claim scope will be a key issue. Practitioners may focus on this term because the defendant could argue that the claims, when read in light of the specification, require the use of motion sensor data (the explicit definition of "path parameters") in addition to image comparison, potentially designing around the patent if their system uses only image analysis.
- Intrinsic Evidence for Interpretation:
- Evidence for a Broader Interpretation: Claim 1 of the '471 patent requires authenticating a user when differences result from "movement of the camera," which could be interpreted broadly to mean any movement that changes the camera's distance to the user, with or without sensor data.
- Evidence for a Narrower Interpretation: The specification explicitly defines path parameters as including data from "an accelerometer, magnetometer, and gyroscope" and states the system uses this data for comparison (’471 Patent, col. 2:12-14, col. 12:35-36). This provides strong evidence for a narrower interpretation requiring the use of non-camera sensor data.
VI. Other Allegations
Indirect Infringement
The complaint alleges that Jumio induces infringement by providing its customers with products and services ("Jumio Accused Instrumentalities") and instructing them on how to use these products in an infringing manner (Compl. ¶¶38-39). The allegations state that Jumio provides "on-screen visual guidance," manuals, and product guides, and configures its system such that users must perform the infringing steps (e.g., moving their device to capture images at different distances) to obtain verification (Compl. ¶¶37, 40).
Willful Infringement
The complaint alleges willful infringement based on Jumio's alleged actual knowledge of the patents-in-suit. This knowledge is predicated on FaceTec having sent a written notice and cease-and-desist letter to Jumio on or about August 23, 2023, which specifically identified the patents and the infringing products (Compl. ¶¶32, 35, 46, 59). The complaint alleges that Jumio's continued infringement after receiving this notice was willful.
VII. Analyst’s Conclusion: Key Questions for the Case
- A core issue will be one of claim scope and prosecution history estoppel: The patents repeatedly emphasize the use of "path parameters" from motion sensors as a key feature. A central question for the court will be whether the asserted claims, which focus on comparing images from different distances, can be infringed by a system that does not use motion sensor data, or if the specification and prosecution history limit the claims to systems that use both imaging and sensor data.
- A key evidentiary question will be one of algorithmic function: Does Jumio's liveness detection software perform the specific comparison recited in the claims—namely, identifying "expected differences" arising from optical distortion and, for the '606 patent, making a determination based on a non-match of biometric data—or does it use a fundamentally different technical method to assess user liveness that falls outside the claim language?
- A third question will relate to prior business relationships: The complaint alleges Jumio was a former licensee of FaceTec's technology. The history of this relationship, the terms of the prior license, and the technical details of the previously licensed product may provide crucial context for claim construction and knowledge supporting the willfulness allegations.