Apple Inc v. Qualcomm Inc

1. Case Identification

• Case #: IPR2019-00186
• Patent #: 9,552,633
• Filed: November 12, 2018
• Petitioner(s): Apple Inc.
• Patent Owner(s): Qualcomm Inc.
• Challenged Claims: 1-7 and 10-28

2. Patent Overview

• Title: Depth Aware Enhancement for Stereo Video
• Brief Description: The ’633 patent discloses methods and systems for enhancing stereoscopic images or video. The core technique involves generating a depth map from a pair of stereo images, using that depth map to identify a specific region for enhancement, and then applying visual enhancements (e.g., sharpening) to the selected region while potentially degrading other regions.

3. Grounds for Unpatentability

Ground 1: Obviousness over Bhaskaran, Sun, and Lynch - Claims 1-7 and 10-28 are obvious over Bhaskaran in view of Sun and Lynch.

• Prior Art Relied Upon: Bhaskaran (Application # 2013/0251281), Sun (Application # 2013/0208093), and Lynch (Application # 2013/0101175).
• Core Argument for this Ground:

  • Prior Art Mapping: Petitioner argued that the combination of the three references taught every limitation of the challenged claims. Bhaskaran provided the foundational system for using a depth map to selectively enhance portions of an image, such as by sharpening a foreground object relative to the background. However, Bhaskaran did not detail how to generate the depth map, inviting the use of any "well-known algorithm." Sun supplied this missing element by teaching a conventional method for generating a depth map by calculating the disparity between corresponding pixels in a left and right stereoscopic image pair.

    The final piece was provided by Lynch, which addressed the key limitation added during the ’633 patent’s prosecution: determining the enhancement region. Lynch taught a technique for defining a "focus area" for enhancement by starting with a user-selected point and then growing a continuous region by identifying surrounding pixels that share a similar depth (i.e., a depth within a certain threshold of the selected point). Petitioner asserted this "seeded region growing" technique from Lynch directly maps onto the claims’ requirements for determining a continuous enhancement region based on a user-selected point and a depth threshold. Dependent claims relating to specific enhancements (e.g., sharpening, blurring, color adjustment) were also taught by Bhaskaran and Lynch.

  • Motivation to Combine: Petitioner contended a person of ordinary skill in the art (POSITA) would have been motivated to combine the references for several reasons. A POSITA starting with Bhaskaran's depth-based enhancement system would naturally look to a well-known reference like Sun to implement the depth map generation, as Bhaskaran itself suggested. Furthermore, a POSITA would have been motivated to incorporate Lynch's user-driven, depth-based segmentation method into the Bhaskaran/Sun system. Lynch offered an efficient and desirable solution for object identification and extraction, allowing a user to precisely select an object of interest for enhancement, which was a common goal in the field of image processing. The combination was presented as a predictable application of known techniques to improve a known system.

  • Expectation of Success: A POSITA would have had a high expectation of success in making this combination. The integration involved applying well-understood and widely used image processing principles—stereopsis from Sun and seeded region growing from Lynch—to a conventional depth-based enhancement framework from Bhaskaran. Because the technologies were highly compatible and the outcome was predictable, success was assured.

4. Key Claim Construction Positions

• Petitioner argued that multiple "means for..." limitations in independent claims 10 and 13 should be construed under 35 U.S.C. § 112(f) as means-plus-function elements.
• The constructions were central to mapping the prior art's disclosures onto the claims. For example, Petitioner proposed that the corresponding structure for the "means for determining a depth map" was a video coder (encoder/decoder) configured to perform disparity calculations, as disclosed in the ’633 patent specification.
• Similarly, the "means for determining an enhancement region" was argued to correspond to a processor configured to execute the specific "region growing" algorithm described in the specification, which involves identifying a continuous portion of an image starting from a user-selected point having a depth within a specified threshold.

5. Relief Requested

• Petitioner requests institution of an inter partes review (IPR) and cancellation of claims 1-7 and 10-28 of the ’633 patent as unpatentable.