PTAB

IPR2024-00604

Amazon.com Inc v. Nokia Technologies Oy

Log In
Key Events
Petition
petition

1. Case Identification

2. Patent Overview

  • Title: Video Encoding and Decoding
  • Brief Description: The ’714 patent discloses methods for improving video compression efficiency by reducing the number of motion vector prediction (MVP) candidates. The claimed invention performs a limited number of comparisons to remove redundant candidates from a candidate list, rather than comparing every possible pair.

3. Grounds for Unpatentability

Ground 1: Claims 1-3, 5-8, 15-17, 19-22, and 29 are obvious over Rusert in view of Karczewicz.

  • Prior Art Relied Upon: Rusert (Application # 2011/0194609) and Karczewicz (Application # 2011/0249721).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued that Rusert, a reference related to the H.264 video coding standard, taught all limitations of the independent claims. Rusert disclosed a method of selecting a list of predicted motion vector (PMV) candidates and removing duplicates by comparing new candidates to a "subset" of previously coded motion vectors. Specifically, Rusert taught selecting a candidate from a set of spatial candidates, determining a subset based on the location of the current block (e.g., via an outward scan within an allowed distance), comparing the candidate's motion information against that subset, and excluding it if redundant, all without comparing every possible pair. Karczewicz was cited to supply the H.265 terminology (e.g., "prediction unit" or PU) that Petitioner contended was an obvious adaptation of Rusert's H.264 "block" terminology.
    • Motivation to Combine: A POSITA would combine these references because Rusert expressly stated its principles could be applied to other coding standards beyond H.264. Karczewicz described H.265 as the successor to H.264, sharing the same block-based prediction concepts. A POSITA would have been motivated to apply Rusert’s established efficiency techniques for de-duplicating candidate lists to the emerging H.265 standard, as described by Karczewicz, to achieve improved coding efficiency.
    • Expectation of Success: A POSITA would have a high expectation of success because applying Rusert’s method to Karczewicz's H.265 framework was a simple substitution of analogous components (H.264 blocks for H.265 PUs) within well-understood video coding architectures. The combination involved applying known techniques to a new standard for a predictable improvement in performance.

Ground 2: Claims 1-8, 15-22, and 29 are obvious over Rusert, Karczewicz, and Lin.

  • Prior Art Relied Upon: Rusert (Application # 2011/0194609), Karczewicz (Application # 2011/0249721), and Lin (Application # 2014/0092981).
  • Core Argument for this Ground:
    • Prior Art Mapping: This ground built upon Ground 1, adding Lin to address the limitations of claim 4. Petitioner argued that Lin taught examining whether a block of pixels is divided into a first and second prediction unit and, if so, excluding a potential spatial MVP candidate if it originates from the second PU. Lin explicitly taught identifying and removing redundant MVP candidates when a Coding Unit (CU) is divided into multiple PUs (e.g., PU1 and PU2), as the motion vector from one PU is not a good predictor for the other.
    • Motivation to Combine: A POSITA would combine Lin with the Rusert/Karczewicz combination because Lin provided a straightforward solution for a known scenario in block-based coding: handling divided blocks. This would further Rusert's goal of reducing the number of candidates to be considered, making the overall process more efficient, particularly in complex video scenes with varied motion.
    • Expectation of Success: Success was expected because Lin's technique for handling partitioned blocks was directly applicable to the H.265 PU-based methods taught by Karczewicz and was a logical extension of the candidate reduction methods taught by Rusert.

Ground 3: Claims 1-8, 15-22, and 29 are obvious over Nakamura in view of WD4.

  • Prior Art Relied Upon: Nakamura (JCTVC-F419, a proposal for the H.265 standard) and WD4 (JCTVC-F803, a working draft of the H.265 standard).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued that Nakamura, a proposal for the H.265 standard, taught a method for reducing the number of candidate comparisons. For both "merge mode" and "MVP mode," Nakamura disclosed deriving spatial candidates (S0 and S1) from a set of spatial neighbors, comparing S1 against a subset containing S0, and removing S1 if it had the same motion information, thereby avoiding a comparison of all possible pairs. WD4, the working draft of the standard that incorporated Nakamura's proposals, provided the necessary context and syntax (e.g., merge_idx) for implementing the method.
    • Motivation to Combine: A POSITA would be motivated to apply Nakamura's teachings to WD4 because that was Nakamura's express purpose. Nakamura was a proposal submitted for inclusion in the H.265 standard, and WD4 was the output of the standards meeting where Nakamura was presented. The references were designed to work together to improve the H.265 standard.
    • Expectation of Success: Success was assured because Nakamura reported actual performance gains from implementing its proposed technique into the H.265 test model software, which was based on the working draft. The combination was not speculative but a demonstrated improvement.

4. Key Claim Construction Positions

  • “spatial motion vector prediction candidate”: Petitioner argued this term means a candidate motion vector obtained from one or more previously-encoded blocks in the current frame.
  • “the block”: Petitioner asserted that, consistent with the prosecution history of a parent patent, "the block" in limitation [1b] refers to the "block of pixels" (i.e., the current block) from limitation [1a], not the block from which the first candidate was obtained. This interpretation is central to how the "subset" of candidates is determined based on the location of the current block.
  • “a subset of … candidates”: Petitioner contended this phrase means one or more candidates, allowing for a comparison of a potential candidate against a subset containing only a single candidate.

5. Arguments Regarding Discretionary Denial

  • Petitioner argued that discretionary denial under Fintiv is unwarranted because a parallel ITC proceeding triggers a mandatory stay of the related district court case, a circumstance under which the PTAB has stated it will not deny institution. Petitioner also argued the petition's merits are compelling.
  • Petitioner contended that denial under §325(d) is also unwarranted because the primary references in Grounds 1 and 2 (Rusert, Karczewicz, Lin) and the WD4 reference in Ground 3 were never cited or considered during prosecution. While Nakamura was cited in a late-submitted IDS, the file history shows no substantive analysis or consideration of it in combination with WD4.

6. Relief Requested

  • Petitioner requests institution of an inter partes review and cancellation of claims 1-8, 15-22, and 29 of the ’714 patent as unpatentable.