Intel Corp v. Fuzzysharp Technologies Inc

1. Case Identification

• Case #: IPR2014-00001
• Patent #: 6,618,047
• Filed: September 30, 2013
• Petitioner(s): Intel Corporation
• Patent Owner(s): Fuzzysharp Technologies, Inc.
• Challenged Claims: 1, 8, 11-13, 46, 57, and 64-67

2. Patent Overview

• Title: Visibility Calculations for 3D Computer Graphics
• Brief Description: The ’047 patent discloses methods for improving the efficiency of 3D computer graphics rendering by reducing the number of required visibility computations. The invention involves pre-processing a 3D scene to determine which surfaces are "always visible" or "always invisible" from a given viewpoint or group of viewpoints, and then ignoring those predetermined surfaces during subsequent, more intensive rendering computations.

3. Grounds for Unpatentability

Ground 1: Claims 1, 8, 12, 13, 46, 57, 64, 65, and 67 are anticipated by Salesin, and Claim 11 is obvious over Salesin in view of Foley.

• Prior Art Relied Upon: Salesin (a 1989 paper titled “The ZZ Buffer: A Simple and Efficient Rendering Algorithm with Reliable Antialiasing”) and Foley (a 1990 textbook titled Computer Graphics: Principles and Practice).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that Salesin disclosed every limitation of the anticipated claims. Salesin taught a method for accelerating hidden-surface determination using a "ZZ-buffer," a data structure that divides an image plane into a grid of rectangular "cells." For each cell, Salesin stored depth data (zmin and zmax) for surfaces within that cell. By comparing a new surface's depth data to the stored data, Salesin's method categorized surfaces as entirely invisible, entirely visible, or potentially visible. As in the ’047 patent, the surfaces determined to be entirely visible or invisible were then ignored (i.e., "thrown away") to reduce computations during the final rendering phase. For claim 11, Petitioner argued Salesin taught all limitations except for surfaces being "arranged in a hierarchy," which was a well-known technique disclosed by Foley.
  • Motivation to Combine (for §103 grounds): Petitioner contended a person of ordinary skill in the art (POSITA) would have been motivated to combine Salesin and Foley. Salesin explicitly stated that it would be "easy to extend the ZZ-buffer algorithm to directly handle objects that are defined recursively, such as... Bezier patches." Foley described these same Bezier patches in the context of hierarchical representations. This direct suggestion in the primary reference, coupled with the renown of the Foley textbook, provided a clear motivation to apply Salesin's efficiency techniques to the hierarchical models described by Foley.
  • Expectation of Success (for §103 grounds): Given Salesin's express statement that its algorithm could be easily extended to handle recursively defined objects like those in Foley, a POSITA would have had a reasonable expectation of success in making the combination.

Ground 2: Claims 1, 8, 11-13, 46, 57, and 64-67 are anticipated or rendered obvious by Airey.

• Prior Art Relied Upon: Airey (a 1990 Ph.D. dissertation titled “Increasing Update Rates in the Building Walkthrough System with Automatic Model-Space Subdivision and Potentially Visible Set Calculation”).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner asserted that Airey disclosed a visibility precomputation method that used the same procedures as the challenged claims. Airey described pre-computing a "potentially visible set" (PVS) of polygons for a given viewpoint region, thereby reducing the number of polygons that needed to be processed during final rendering. This process involved projecting polygons onto a plane and using a "sampling grid" to determine occlusion. If a polygon was found to be occluded (invisible), further processing on it was "aborted," thus meeting the "ignoring" limitation. While the main body of Airey's method focused on identifying invisible surfaces, Petitioner argued that other sections of the dissertation discussed the concept of determining "completely visible" surfaces, making the claims either anticipated or, at a minimum, obvious over Airey alone.
  • Motivation to Combine (for §103 grounds): Petitioner argued that even if Airey's teachings on "always visible" and "always invisible" surfaces were in different sections, a POSITA would have been motivated to combine them. Airey explicitly recapitulated the known tripartite visibility relationship (completely visible, completely invisible, or partially visible) and its benefits for computational savings. This discussion within Airey itself suggested applying the "always visible" classification to its PVS method to achieve even greater efficiency, rendering the combination obvious.

Ground 3: All challenged claims are anticipated by the patent’s own PCT application counterpart.

• Prior Art Relied Upon: The Australian PCT application counterpart to the ’047 patent (No. PCT/AU92/00302), published January 7, 1993.
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that the PCT application was virtually identical in text and drawings to the ’047 patent and therefore disclosed every element of the challenged claims. The core of this ground was a legal argument, not a technical one. Petitioner contended the ’047 patent was not entitled to its claimed June 28, 1991 priority date. The applicant failed to file the U.S. national-stage application within the required 30-month deadline, causing it to be abandoned. The applicant later revived the application by asserting the delay was "unintentional." However, Petitioner argued that the governing statute, 35 U.S.C. §371(d), required a showing that the delay was "unavoidable"—a much higher standard. Because the examiner improperly revived the application under the more lenient "unintentional" standard, the revival was legally ineffective. Therefore, the ’047 patent's effective filing date was its actual U.S. filing date of June 1, 1994, making the PCT application published in 1993 statutory prior art under 35 U.S.C. §102(b) that anticipated all claims.

4. Key Claim Construction Positions

• "determining which of said at least one of 3-D surfaces or their sub-elements is always invisible or always visible" (claims 1, 8): This was the sole disputed construction.
  • Petitioner's Position: The term required determining both which surfaces are totally hidden and which are totally visible.
  • Patent Owner's Position (in litigation): The term only required making one or the other of those determinations (i.e., determining which surfaces are hidden or which are visible).
  • Relevance: Petitioner argued the dispute was irrelevant to patentability because all cited prior art references disclosed determining both categories of surfaces, meaning the claims were invalid even under the Petitioner's narrower construction.

5. Relief Requested

• Petitioner requests institution of an inter partes review and cancellation of claims 1, 8, 11-13, 46, 57, and 64-67 of the ’047 patent as unpatentable.