PTAB

IPR2014-00804

TMM Inc v. Dimension Inc

Log In
Key Events
Petition
petition Intelligence

1. Case Identification

2. Patent Overview

  • Title: Methods and Systems for Up-Scaling a Standard Definition (SD) Video to High Definition (HD) Quality
  • Brief Description: The ’053 patent relates to a computerized system for up-scaling a lower-resolution source video to a higher-resolution output video. The system primarily uses a block-based method described as "fractal zooming" to replace individual source pixels with a plurality of best-matching replacement pixels found elsewhere in the source video.

3. Grounds for Unpatentability

Ground 1: Obviousness of Core Fractal Zooming Method - Claims 1-4, 7, 13-21, and 27-30 are obvious over Cubillo, Ebrahimi, and Gonzalez.

  • Prior Art Relied Upon: Cubillo (Patent 6,141,017), Ebrahimi (a 2008 Ph.D. Thesis), and Gonzalez (a 2008 digital imaging textbook).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued that the core fractal zooming process of independent claim 1 was a predictable combination of known elements. Cubillo was asserted to teach a block-based fractal zooming system for up-scaling video images by replacing domain blocks with center data from best-matching range blocks. To address the claim limitation of zooming "upon each individual source pixel," Petitioner cited Ebrahimi, which explicitly taught a fractal zooming process performed on individual pixels. For the remaining conventional video processing steps recited in claim 1 (color space conversion, low-pass filtering, down-sampling, and outputting), Petitioner relied on Gonzalez as a standard textbook teaching that these are fundamental techniques in image processing.
    • Motivation to Combine: Petitioner contended that a POSITA would combine Cubillo and Ebrahimi as an obvious design choice, modifying Cubillo's block-based approach to the pixel-based approach of Ebrahimi to achieve clearer images with reduced computational complexity. A POSITA would further combine these teachings with the conventional processing steps from Gonzalez to optimize the video processing system, a standard practice in the field.
    • Expectation of Success: Petitioner asserted a high expectation of success because the combination involved applying known techniques (pixel-based zooming, filtering, color conversion) for their intended purposes, yielding predictable results in video enhancement.

Ground 2: Obviousness with De-Interlacing - Claims 5 and 6 are obvious over Cubillo, Ebrahimi, and Gonzalez, in further view of Thompson.

  • Prior Art Relied Upon: Cubillo (Patent 6,141,017), Ebrahimi (a 2008 Ph.D. Thesis), Gonzalez (a 2008 digital imaging textbook), and Thompson (Patent 7,027,099).
  • Core Argument for this Ground:
    • Prior Art Mapping: This ground built upon the core combination of Ground 1. To meet the limitations of claims 5 and 6, which required the step of "de-interlacing the source video," Petitioner introduced Thompson. Thompson was alleged to disclose a well-known method for de-interlace processing, which converts interlaced video to progressive scan video by interpolating pixel data for blank rows, thereby improving video resolution.
    • Motivation to Combine: A POSITA would be motivated to add Thompson's de-interlacing step as a pre-processing stage to the fractal zooming algorithm to improve the overall look and clarity of the final up-scaled video, a common goal in video processing.
    • Expectation of Success: Petitioner argued success would be predictable, as de-interlacing was a standard, self-contained video improvement technique that could be readily applied before other processing steps without disrupting their function.

Ground 4: Obviousness with Per-Component Processing - Claims 22, 23, and 25 are obvious over Cubillo, Ebrahimi, and Gonzalez, in further view of Ishizaka.

  • Prior Art Relied Upon: Cubillo (Patent 6,141,017), Ebrahimi (a 2008 Ph.D. Thesis), Gonzalez (a 2008 digital imaging textbook), and Ishizaka (Patent 7,046,862).

  • Core Argument for this Ground:

    • Prior Art Mapping: This ground added Ishizaka to the core combination to address claims related to processing color components. Ishizaka was cited for teaching a fractal method for increasing image resolution where each color component (e.g., R, G, B) is enlarged separately (for claim 22) or simultaneously by creating a "quasi-color image" (for claim 23). For claim 25's limitation of comparing only luminance values, Ishizaka was cited for teaching the creation of a quasi-gray image substantially equal to illuminance components for matching purposes.
    • Motivation to Combine: Petitioner argued a POSITA would be motivated to adopt Ishizaka's methods to reduce the complexity of the video processing algorithm and shorten processing time, particularly when processing color video. Applying zooming to luminance values alone was argued to be an obvious optimization to exploit the human eye's heightened sensitivity to luminance over chrominance.
    • Expectation of Success: The combination was asserted to be predictable, as applying the zooming algorithm to individual or combined color components was a known, alternative way to handle color data.

  • Additional Grounds: Petitioner asserted additional obviousness challenges. Ground 3 added the Focus Guide (a data sheet) to the core combination to teach trimming video frames and confirming color space. Ground 5 added Lu (a 1997 publication) to the Ground 4 combination to teach comparing both luminance and chrominance values and weighing luminance higher.

4. Relief Requested

  • Petitioner requests institution of an inter partes review and cancellation of claims 1-30 of Patent 8,639,053 as unpatentable.