Samsung Electronics Co Ltd v. M & K Holdings Inc

1. Case Identification

• Patent #: 9,197,896
• Filed: February 26, 2018
• Petitioner(s): Samsung Electronics Co., Ltd.
• Patent Owner(s): M & K Holdings Inc.
• Challenged Claims: 1-5

2. Patent Overview

• Title: Method and System for Intracoding in Video Encoding
• Brief Description: The ’896 patent relates to a video encoding method using intra-prediction. The method involves generating a prediction block, creating a residual block, transforming and quantizing the residual block into coefficients, and then adaptively scanning and entropy-coding those coefficients based on the prediction mode.

3. Grounds for Unpatentability

Ground 1: Obviousness over Budagavi and Sole - Claim 1 is obvious over Budagavi in view of Sole.

• Prior Art Relied Upon: Budagavi (Application # 2011-0170594) and Sole (JCTVC-E335, a March 2011 contribution to the HEVC standard).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that Budagavi taught the fundamental elements of claim 1, including an image encoding method with an intra-predictor, a transform/quantization unit to create a quantized block from a residual block, and an entropy coder. However, Budagavi only generally disclosed scanning the quantized coefficients. Petitioner asserted that Sole, a proposal for the HEVC standard, explicitly taught the missing elements: dividing a quantized block (specifically an 8x8 block) into a plurality of subsets (e.g., 4x4 sub-blocks) and scanning both the subsets and the coefficients within them according to a scan pattern (e.g., zig-zag, horizontal, or vertical) that is determined by the intra-prediction mode.
  • Motivation to Combine: Petitioner contended a person of ordinary skill in the art (POSITA) would combine these references to improve the efficiency of Budagavi’s encoding process. Budagavi itself suggested that its methods were applicable to the HEVC standard and even referenced proposals to the same working group that produced Sole. Therefore, a POSITA looking to implement a more efficient scanning technique than Budagavi’s basic raster-scan suggestion would naturally look to contemporaneous HEVC proposals like Sole, which were aimed at improving coding efficiency for the same technical purpose.
  • Expectation of Success: A POSITA would have reasonably expected success in combining these teachings. The integration involved applying a known, more advanced scanning technique (Sole) to a conventional video encoding framework (Budagavi) to achieve the predictable result of more efficient compression, a primary goal in the field.

Ground 2: Obviousness over Budagavi, Sole, and WD3-v1 - Claims 2-5 are obvious over Budagavi and Sole in view of WD3-v1.

• Prior Art Relied Upon: Budagavi (Application # 2011-0170594), Sole (JCTVC-E335), and WD3-v1 (JCTVC-E603, a March 2011 working draft of the HEVC standard).
• Core Argument for this Ground:
  • Prior Art Mapping: This ground builds upon the combination of Budagavi and Sole from Ground 1. Petitioner argued that WD3-v1, a working draft of the HEVC standard, provided the specific additional limitations required by dependent claims 2-5. For claim 2, WD3-v1 taught coding an intra-prediction mode based on neighboring upper and left predictors and, if a neighbor is unavailable, setting its mode to a default DC mode. For claims 3 and 4, WD3-v1 provided specific mappings, teaching the use of a horizontal scan pattern for a vertical prediction mode (claim 3) and a vertical scan pattern for a horizontal prediction mode (claim 4). For claim 5, WD3-v1 taught a method for generating reference pixels when they are unavailable, for example at the border of a frame.
  • Motivation to Combine: Petitioner asserted a POSITA, having combined Budagavi and Sole to create a more efficient scanning process, would consult a working draft of the same standard (WD3-v1) to implement robust solutions for common scenarios not fully detailed in the primary references. These include handling unavailable predictors (claim 2), defining specific scan-pattern-to-prediction-mode relationships for improved efficiency (claims 3-4), and managing unavailable reference pixels at image boundaries (claim 5). Incorporating these established techniques from the developing HEVC standard would be a logical step to create a more complete and efficient encoder.

4. Key Claim Construction Positions

• Petitioner argued that terms in the challenged method claims reciting structural components (e.g., "transform/quantization unit," "intra predictor," "entropy coder") should not be construed as means-plus-function limitations under §112, 6th paragraph.
• The argument was based on prosecution history estoppel. During prosecution, the Examiner issued a rejection because the original apparatus claims invoked §112, 6th paragraph, but the specification lacked clear corresponding structure. In response, the applicant amended the claims from apparatus claims to method claims and removed "configured to" language, thereby, in Petitioner's view, clearly and unmistakably disavowing any means-plus-function interpretation to secure allowance.

5. Key Technical Contentions (Beyond Claim Construction)

• Petitioner dedicated significant argument to establishing that the challenged claims were not entitled to the benefit of their earliest claimed priority date (a Korean application from April 2010).
• The core contention was that the Korean priority application failed to provide adequate written description support for a key limitation of claim 1: "when a size of the quantized block is equal to 8x8, the quantized coefficients... are divided into a plurality of subsets, and the... subsets are scanned according to a scan pattern determined by the intra prediction mode."
• By establishing a later effective filing date (April 22, 2011), Petitioner positioned the Sole and WD3-v1 references—both published before this date—as valid prior art under pre-AIA §102(a).

6. Relief Requested

• Petitioner requested institution of an inter partes review and cancellation of claims 1-5 of the ’896 patent as unpatentable.