PTAB

IPR2021-00705

Unified Patents LLC v. Sovereign Peak Ventures LLC

Log In
Key Events
Petition

1. Case Identification

2. Patent Overview

  • Title: Image Decoding Method, Image Decoding Apparatus, Image Coding Method, And Image Coding Apparatus
  • Brief Description: The ’457 patent discloses an image decoding method intended to solve processing order dependencies in video codecs, such as the H.264 standard, to enable parallel decoding of video data. The method involves pre-decoding certain syntax elements from a coded stream to generate parameters that are then used for parallel decoding of multiple macroblock groups.

3. Grounds for Unpatentability

Ground 1: Obviousness over Pearson and H.264 - Claims 7-8 are obvious over Pearson in view of H.264.

  • Prior Art Relied Upon: Pearson (Patent 7,983,343) and H.264 (Advanced Video Coding Standard, Rec. ITU-T H.264, Mar. 2005).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued that Pearson taught a method to speed up Context-Based Adaptive Binary Arithmetic Coding (CABAC) decoding, which the ’457 patent identified as a type of variable length decoding. Pearson’s system pre-decodes syntax elements from a bitstream to generate "neighbor macroblock data" (e.g., Mb_qp_delta, a quantization parameter), which corresponds to the claimed "block decoding information." This information is then used by the CABAC decoder to decode multiple macroblock groups (the current macroblock and its neighbors) in parallel, meeting the limitations of independent claim 7. Petitioner asserted that the claimed step of "re-executing the variable length decoding" was met because Pearson’s decoder first performs a partial decode to obtain the neighbor data and then uses that data to perform a full decode of the macroblocks in parallel to determine prediction modes. H.264 was cited to supply the background understanding that CABAC decoding is performed on a "block group-by-block group basis" (i.e., on slices, which are pluralities of macroblocks). Dependent claim 8, which requires the block decoding information to include a quantization parameter, was allegedly met by Pearson’s disclosure of generating "Mb_qp_delta" as part of the neighbor data.
    • Motivation to Combine: Petitioner asserted that a person of ordinary skill in the art (POSITA) would be motivated to combine these references because Pearson explicitly presents its system as an improvement on conventional H.264 codecs. Pearson repeatedly referenced the H.264 standard, stated its decoder "may be an H.264 decoder," and incorporated relevant portions of the standard by reference, providing an explicit motivation for the combination.
    • Expectation of Success: A POSITA would have had a reasonable expectation of success in applying the parallel processing teachings of Pearson to a standard H.264 decoding framework, as Pearson’s invention was specifically designed to improve upon and operate within that framework.

Ground 2: Obviousness over Chen and H.264 - Claim 7 is obvious over Chen in view of H.264.

  • Prior Art Relied Upon: Chen (Jian-Wen Chen et al., “A High-Performance Hardwired CABAC Decoder,” 2007 IEEE Conference Publication) and H.264.
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued that Chen disclosed a parallel decoding architecture for H.264/AVC to reduce clock cycles. Chen's architecture used a Variable Length Coding (VLC) parameter decoder to first decode "slice-level information" into a parameter memory. This pre-decoded information corresponds to the claimed "block decoding information." The main CABAC decoder then used these parameters from memory to process the stream data for multiple macroblocks in parallel. This two-stage process—a first VLC decoding to generate parameters followed by a second, parallel CABAC decoding using those parameters—was argued to meet the limitations of claim 7, including the "re-executing" step. The parallel decoding process disclosed in Chen was used to determine syntax elements like the Coded_block_flag, which is necessary for determining the prediction mode. H.264 was again used to provide the context that such decoding systems operate on slices, which constitute "block groups."
    • Motivation to Combine: Petitioner argued the motivation to combine was explicit, as Chen's paper presented a "high-performance hardwired context-based adaptive binary arithmetic decoder (CABAD) for H.264/AVC." The entire system was designed as an implementation of and improvement for the H.264 standard, providing a clear reason for a POSITA to read its teachings in conjunction with the standard itself.
    • Expectation of Success: A POSITA would have expected success in combining Chen with H.264 because Chen’s architecture was purpose-built for H.264 decoding, and the paper demonstrated a successful FPGA prototype of the system.

4. Key Claim Construction Positions

  • Petitioner argued for a construction of the term "Block Decoding Information" based on the specification and prosecution history.
  • The proposed construction was "a parameter necessary for decoding other macroblock groups." Petitioner contended this term should be interpreted broadly and not be limited to the specification's example of a quantization parameter, a position supported by the doctrine of claim differentiation, as dependent claim 8 separately added that limitation.

5. Arguments Regarding Discretionary Denial

  • Petitioner presented substantial arguments that discretionary denial under either 35 U.S.C. §325(d) or §314(a) would be inappropriate.
  • §325(d) (Advanced Bionics): Petitioner argued that neither the primary references (Pearson and Chen) nor the asserted combinations were ever presented to or considered by the Examiner during prosecution. While H.264 was cited in an IDS, it was never substantively discussed or used in a rejection, meaning the core of Petitioner's arguments was new.
  • §314(a) (Fintiv): Petitioner argued that all six Fintiv factors weighed against denial. Key arguments included that Petitioner was not a party to the parallel district court litigation involving TCL; the Board would almost certainly issue a final written decision before a trial would occur in that case; investment in the court case was minimal; and the merits of the petition were exceptionally strong, favoring institution to serve the public interest in re-evaluating patents that should not have issued.

6. Relief Requested

  • Petitioner requested institution of an inter partes review and cancellation of claims 7-8 of Patent 9,042,457 as unpatentable under 35 U.S.C. §103.