Apple Inc v. Corephotonics Ltd

1. Case Identification

• Case #: IPR2020-00905
• Patent #: 10,225,479
• Filed: May 6, 2020
• Petitioner(s): Apple Inc.
• Patent Owner(s): Corephotonics Ltd.
• Challenged Claims: 1-16, 18, 23-38, and 40

2. Patent Overview

• Title: Dual Aperture Zoom Digital Camera
• Brief Description: The ’479 patent describes a dual-aperture digital camera system, such as for a cell phone, that includes separate wide-angle and telephoto lens/sensor modules. The system is configured to process and fuse images from both modules to produce a single output image with an enhanced depth of field.

3. Grounds for Unpatentability

Ground 1: Obviousness over Parulski and Konno - Claims 1, 10-14, 16, 18, 23, 32-36, 38, and 40 are obvious over Parulski in view of Konno.

• Prior Art Relied Upon: Parulski (Patent 7,859,588) and Konno (JP Pub. No. 2013-106289).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that Parulski taught the core system of the ’479 patent: a dual-lens cell phone camera with wide and telephoto lenses that captures two images at different focus positions and combines them to create a fused image with a broadened depth of field. Parulski disclosed that focused portions of the telephoto image are combined with the wide image. Konno was argued to supply specific teachings of a compact, fixed-focal-length, dual-lens system suitable for a cell phone, including a telephoto lens with an effective focal length to total track length ratio (EFLT/TTLT) greater than 1, as required by independent claims 1 and 23. Konno also taught features of dependent claims, such as image sensors with identical pixel counts (claim 10) and different F-numbers for the lenses (claim 14).
  • Motivation to Combine: A Person of Ordinary Skill in the Art (POSITA) would combine Konno’s lens system with Parulski’s camera framework. Parulski described a general dual-lens system but provided no specific lens prescription, motivating a POSITA to seek a suitable, pre-existing design. Konno provided a compact, high-performance, and low-cost dual-lens system explicitly designed for mobile devices, directly addressing the design constraints (e.g., small thickness) emphasized in Parulski.
  • Expectation of Success: A POSITA would have a reasonable expectation of success because it involved incorporating a known type of compact dual-lens system (Konno) into a known camera architecture (Parulski) to achieve the predictable result of a functional dual-lens camera with image fusion capabilities.

Ground 2: Obviousness over Parulski, Konno, and Szeliski - Claims 2-4 and 24-26 are obvious over Parulski, Konno, and Szeliski.

• Prior Art Relied Upon: Parulski (Patent 7,859,588), Konno (JP Pub. No. 2013-106289), and Szeliski (a 2011 computer vision textbook).
• Core Argument for this Ground:
  • Prior Art Mapping: This ground built upon the combination of Parulski and Konno. The additional limitations in claims 2-4 required the camera controller to perform image rectification by aligning the wide and telephoto images to an epipolar line, perform mapping to create a registration map, and resample the telephoto image based on that map. Petitioner asserted that Szeliski, a standard textbook on computer vision, explicitly taught these techniques. Szeliski explained that rectifying stereo images along an epipolar line is a well-known, foundational step that makes subsequent pixel matching and registration map generation more efficient and accurate.
  • Motivation to Combine: A POSITA implementing Parulski’s range-mapping and image fusion process would be motivated to incorporate Szeliski’s rectification step to improve the efficiency and reliability of the algorithm. Since both Parulski and Szeliski addressed the problem of calculating distance information from stereo images, and Szeliski provided a known method to improve that calculation, the combination was presented as a straightforward optimization.
  • Expectation of Success: Success was expected because applying a standard, efficiency-improving software algorithm (Szeliski) to an existing image processing framework (Parulski) was a predictable and well-understood engineering task.

Ground 3: Obviousness over Parulski, Konno, Szeliski, and Segall - Claims 5-9 and 27-31 are obvious over Parulski, Konno, Szeliski, and Segall.

• Prior Art Relied Upon: Parulski (Patent 7,859,588), Konno (JP Pub. No. 2013-106289), Szeliski (a 2011 computer vision textbook), and Segall (Patent 8,406,569).
• Core Argument for this Ground:
  • Prior Art Mapping: This ground added the teachings of Segall to address limitations in claims 5 and 6, which required detecting registration errors between the images and, if an error is detected, choosing the wide image pixel values for the output. Petitioner argued that while Parulski taught image fusion, it did not specify how to handle misregistration errors. Segall was cited for teaching a two-phase image processing method (registration and fusion) that included a "mismatch detector" to identify registration errors and "selectively exclude" pixels from enhancement frames (the telephoto image) that were not accurately aligned with the reference frame (the wide image). This exclusion would result in using only the reference (wide) pixel values, directly mapping to the claim limitation.
  • Motivation to Combine: A POSITA implementing the fusion process of Parulski would encounter the common problem of registration errors. Segall taught a known solution for this exact problem. Therefore, a POSITA would be motivated to incorporate Segall’s error-handling technique into Parulski’s system to improve the quality and robustness of the final fused image.
  • Expectation of Success: The combination was a predictable application of a known error-handling software module (Segall) to a known image fusion process (Parulski) to solve an expected problem.
• Additional Grounds: Petitioner asserted an additional obviousness challenge (Ground 4) for claims 15 and 37 based on the combination of Parulski, Konno, and Stein (Patent 8,908,041). Stein was used to teach synchronizing the scanning of CMOS image sensors with rolling shutters to ensure matching fields of view are scanned at the same time, a technique to improve stereo image processing.

4. Key Claim Construction Positions

• "fused image with a point of view (POV) of the Wide camera" (claims 1 and 23): Petitioner argued this term should be construed to mean "a fused image that maintains the Wide camera's field of view or the Wide camera's position." This construction was based on the specification's description that the output image could retain either the "Wide perspective POV" (field of view) or the "Wide position POV" (camera position). Petitioner contended this construction was important because it encompassed prior art methods, like Parulski's, where fusing a telephoto image portion onto a wide image inherently maintains the overall field of view of the wide image.

5. Key Technical Contentions (Beyond Claim Construction)

• Correctability of Konno's Lens Design: Petitioner's expert contended that the telephoto lens design in Konno's "Example 2" contained a minor, physically impossible overlap (0.5 microns) between two lens elements as disclosed. The expert argued a POSITA would have easily recognized this as a design artifact and corrected it by adjusting one element's position, resulting in a functional lens with comparable performance. This contention was critical to establishing Konno as an enabling prior art reference for the claimed telephoto lens structure.

6. Relief Requested

• Petitioner requested institution of an inter partes review and cancellation of claims 1-16, 18, 23-38, and 40 of the ’479 patent as unpatentable.