PTAB
IPR2019-00720
Unified Patents Inc. v. Velos Media, LLC
1. Case Identification
- Case #: IPR2019-00720
- Patent #: 9,538,205
- Filed: March 1, 2019
- Petitioner(s): Unified Patents Inc.
- Patent Owner(s): Velos Media, LLC
- Challenged Claims: 1-5
2. Patent Overview
- Title: Arithmetic Coding and Decoding Devices for Images
- Brief Description: The ’205 patent relates to arithmetic coding and decoding devices for images that use an arithmetic compression algorithm. The technology focuses on using "derivation patterns" to group sub-blocks of image data into contexts for efficient compression, allegedly improving upon prior art by adding a third context group for cases with an intermediate probability of non-zero transform coefficients.
3. Grounds for Unpatentability
Ground 1: Claims 1-5 are obvious over Ji in view of Kumakura.
- Prior Art Relied Upon: Ji (Application # 2013/0188726) and Kumakura (JCTVC-I0296, a 2012 technical contribution to the Joint Collaborative Team on Video Coding).
- Core Argument for this Ground:
- Prior Art Mapping: Petitioner argued that Ji teaches a video coding system using three distinct context sets or regions to encode transform coefficients. Kumakura teaches a method for deriving context maps that analyzes the right and lower adjacent sub-blocks for significant coefficient flags. Specifically, when both adjacent blocks lack significant coefficients, Kumakura defines two context groups using a diagonal boundary. The combination of Ji's three-context framework with Kumakura's adjacent-block analysis was asserted to teach the limitations of the challenged claims.
- Motivation to Combine: A person of ordinary skill in the art (POSITA) would combine Kumakura's efficient adjacent-block analysis with Ji's three-context framework. The motivation would be to improve throughput and reduce time complexity, which are well-known objectives in the field of video compression.
- Expectation of Success: A POSITA would have a reasonable expectation of success because the combination was presented as a predictable extension of processes already disclosed in Ji, which itself considers adjacent block flags as part of its context model.
Ground 2: Claims 1, 2, and 4 are obvious over Kumakura in view of Kung.
- Prior Art Relied Upon: Kumakura (JCTVC-I0296) and Kung (JCTVC-I0372, another 2012 technical contribution).
- Core Argument for this Ground:
- Prior Art Mapping: Petitioner asserted that Kumakura teaches deriving context maps based on adjacent blocks, using two context groups when those adjacent blocks lack significant coefficients. Kung was argued to disclose an improved system with three context "zones": one for the DC coefficient (where coordinate sum is zero), a second for coefficients below a threshold, and a third for those at or above the threshold. This three-zone approach provides a more granular context model than Kumakura's two-group system.
- Motivation to Combine: A POSITA would be motivated to modify Kumakura's two-context system with Kung's more granular three-context system. This combination would decrease encoding and decoding time with negligible compression loss, thereby providing a more optimal quantization strategy, a clear goal for anyone working in the field.
- Expectation of Success: The modification would be a straightforward extension of Kumakura's existing diagonal boundary framework. A POSITA would simply apply Kung's logic to define a third context region, leading to predictable improvements in efficiency.
Ground 3: Claims 3 and 5 are obvious over Kumakura in view of Kung and Ji.
- Prior Art Relied Upon: Kumakura (JCTVC-I0296), Kung (JCTVC-I0372), and Ji (Application # 2013/0188726).
- Core Argument for this Ground:
- Prior Art Mapping: This ground targets apparatus claims 3 and 5 by building upon the previous grounds. Petitioner argued that the combination of Kumakura and Kung provides an improved three-context quantization method. Ji, in turn, provides the standard, necessary components of a complete encoding and decoding pipeline, such as an inverse transform processor, spatial compensator, and entropy encoder, which are not the focus of Kumakura or Kung.
- Motivation to Combine: A POSITA would be motivated to implement the improved, efficient quantization scheme derived from the Kumakura/Kung combination into the comprehensive and well-understood pipeline architecture disclosed by Ji. The goal would be to create a complete, fully functional, and more efficient image coding and decoding apparatus.
- Expectation of Success: Integrating an improved functional module (the context mapping from Kumakura/Kung) into a standard, known system architecture (the pipeline from Ji) is a predictable design choice. A POSITA would expect the resulting system to function as intended and exhibit the efficiency gains from the improved module.
4. Key Claim Construction Positions
- Petitioner argued that several "unit" limitations in the claims invoke means-plus-function treatment under pre-AIA 35 U.S.C. §112, ¶6. These include the "sub-block coefficient presence/absence flag decoding unit" (claim 1), "context index deriving unit" (claims 1, 2, 4), "inverse frequency transform unit" (claim 3), and "transform coefficient generating unit" (claim 5).
- For each of these limitations, Petitioner proposed that the claimed function is as recited in the claim and the corresponding structure is a CPU executing specific software algorithms. These algorithms are detailed in the ’205 patent's specification and illustrated in its flowcharts and figures (e.g., Figures 7, 16, 40, 43). This construction was asserted to be critical for mapping the algorithmic steps disclosed in the prior art to the claimed "units."
- Petitioner also proposed that the terms "transform coefficient presence/absence flag" and "sub-block coefficient presence/absence flag" should be construed as corresponding to the specific syntax elements
significant_coeff_flag
andsignificant_coeff_group_flag
, respectively, as explicitly defined in the ’205 patent specification.
5. Arguments Regarding Discretionary Denial
- Petitioner argued that discretionary denial under 35 U.S.C. §325(d) would be inappropriate because the core arguments and prior art combinations were not previously considered by the USPTO.
- It was asserted that the primary prior art references, Ji and Kung, were never cited or considered during the original prosecution.
- While Kumakura was submitted by the applicant during prosecution, Petitioner contended that the Examiner did not substantively address it in a rejection or provide any analysis of its teachings. Therefore, the invalidity grounds presented in the petition were argued to be new and not cumulative to what the Examiner reviewed.
6. Relief Requested
- Petitioner requested the institution of an inter partes review and the cancellation of claims 1-5 of Patent 9,538,205 as unpatentable.