Align Technology Inc v. 3Shape AS

1. Case Identification

• Case #: IPR2021-01310
• Patent #: 10,383,711
• Filed: July 23, 2021
• Petitioner(s): Align Technology, Inc.
• Patent Owner(s): 3Shape As
• Challenged Claims: 1-2, 9-11, 23-24, 40, and 42

2. Patent Overview

• Title: Focus Scanning Apparatus Recording Color
• Brief Description: The ’711 patent relates to a three-dimensional (3D) scanner system for recording both surface geometry and surface color of an object. The system uses a multichromatic light source, a pattern generator, and a single color image sensor with a color filter array to derive geometry information by identifying the focus plane of maximum contrast from a series of 2D images.

3. Grounds for Unpatentability

Ground 1: Claims 1-2, 9-11, 40, and 42 are obvious over Fisker.

• Prior Art Relied Upon: Fisker (Application # 2012/0092461).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that Fisker taught a focus-scanning system for capturing both 3D geometry and color. While Fisker described a preferred embodiment using sequential illumination, it also disclosed an alternative configuration using a multichromatic light source (e.g., white light) and a color image sensor with a color filter array (e.g., a Bayer filter), which mapped to the system of claim 1. Critically, Fisker suggested that to reduce the amount of data processed, geometry could be determined using only one color, while color information would be derived from all colors. Petitioner contended this directly taught deriving geometry from a "first set" of image sensor pixels (those corresponding to one color filter) and color from a "second set" (those corresponding to all color filters), making the sets different as required by the claim.
  • Motivation to Combine: A POSITA would have been motivated to implement Fisker’s alternative configuration (multichromatic light source with a color filter array sensor) because it was a well-known method that would decrease scan time and reduce system complexity compared to Fisker’s sequential illumination embodiment. Furthermore, a POSITA would have been motivated to limit geometry calculations to a single color channel, as Fisker suggested, to reduce data processing and provide a more uniform dataset for deriving surface geometry.
  • Expectation of Success: Petitioner asserted a POSITA would have had a reasonable expectation of success because the components and methods described, including the use of a color filter array and demosaicing (a known technique for "combining and/or weighing" color data as taught by Fisker), were conventional and their use in 3D scanning was routine and predictable.

Ground 2: Claims 23 and 24 are obvious over Fisker in view of Yamada.

• Prior Art Relied Upon: Fisker (Application # 2012/0092461) and Yamada (Application # 2007/0035641).
• Core Argument for this Ground:
  • Prior Art Mapping: This ground built upon the Fisker reference to address claims 23 and 24, which recite detecting and mitigating errors from saturated pixels. Petitioner argued that Fisker provided the base scanner system capable of deriving surface color. Yamada taught a method for correcting "false color" errors caused by white-saturated pixels in 2D images. Yamada's method involved detecting a saturated pixel and then correcting the color values of surrounding pixels by interpolating from neighboring, non-saturated pixels. This process effectively excluded the saturated pixel from the color calculation, which Petitioner argued was equivalent to "assigning a low weight" (specifically, a zero weight) to the saturated pixel's color information, as recited in claim 24.
  • Motivation to Combine: A POSITA would have been motivated to combine Yamada’s method with Fisker's system because pixel saturation was a well-known problem in digital imaging. Implementing Yamada’s routine error-correction technique would predictably and efficiently produce higher-quality color images without requiring complex modifications to the optical system.
  • Expectation of Success: Success was expected because Yamada’s method was a conventional solution to a known problem, and its application to the color data generated by Fisker’s standard imaging components would have been straightforward for a POSITA.

Ground 3: Claims 1, 9-11, 40, and 42 are obvious over Colonna de Lega and Min.

• Prior Art Relied Upon: Colonna de Lega (WO 2013/105922) and Min (Application # 2011/0085063).

• Core Argument for this Ground:

  • Prior Art Mapping: Petitioner presented this as an alternative combination for the main set of challenged claims. Colonna de Lega taught a scanner system that captured surface geometry and color using a broadband light source and a color image sensor (e.g., a single CCD with a Bayer filter). Min taught a color image sensor with a specialized filter array that included two distinct sets of filters: one set for color information (e.g., red, green, blue filters) and another set for depth information (e.g., white filters). Petitioner argued that combining Min's sensor design into Colonna de Lega’s system directly provided for deriving geometry from a "first set" of pixels (those with white filters) and color from a "second, different set" of pixels (those with RGB filters). Min also explicitly taught using a demosaicing process to obtain full color information.
  • Motivation to Combine: A POSITA would combine Min’s filter array into a scanner like Colonna de Lega’s to enable simultaneous 3D and color imaging from a single image capture. Min's design was advantageous as it was taught to maintain a good signal-to-noise ratio for accurate depth information while still allowing for color capture, providing a clear reason for its adoption.
  • Expectation of Success: The principles of color imaging and the components involved were well known. Implementing Min's specialized but well-defined filter array into Colonna de Lega's system would have involved applying known techniques with predictable results.

• Additional Grounds: Petitioner asserted an additional obviousness challenge against claims 23 and 24 over Colonna de Lega, Min, and Yamada, relying on the same saturation-correction rationale provided by Yamada as described in Ground 2.

4. Key Claim Construction Positions

• "low weight" (claim 24): Petitioner argued that whatever the term's upper bounds, assigning a zero weight to the color information of a saturated pixel, as taught by Yamada's error-correction method, would qualify as a "low weight."
• "data processing system" (claim 1): Petitioner noted its contention that the term is a means-plus-function term but applied the Patent Owner’s litigation position (that it is not) for purposes of the petition, arguing the prior art disclosed the corresponding structure of processors and algorithms for performing the recited functions.

5. Arguments Regarding Discretionary Denial

• §325(d) (Previously Presented Art/Arguments): Petitioner argued against discretionary denial because the petition presented materially different art and arguments than those before the Examiner during prosecution. Specifically, the Examiner did not consider Colonna de Lega or Min. While Fisker was cited, the Examiner did not have the benefit of expert testimony explaining how Fisker’s disclosures suggested using different pixel sets or how demosaicing was a conventional technique consistent with its teachings.
• §314(a) (Parallel Litigation / Fintiv Factors): Petitioner argued denial under Fintiv would be inappropriate. It asserted that the parallel district court litigation was in its early stages with no trial date set, and that a stay of that litigation would be sought, favoring institution. Petitioner also argued that the strength of the prior art and the merits of the unpatentability challenges weighed strongly against discretionary denial.

6. Relief Requested

• Petitioner requests institution of an inter partes review (IPR) and cancellation of claims 1-2, 9-11, 23-24, 40, and 42 of the ’711 patent as unpatentable.