Google LLC v. Cellular South Inc

1. Case Identification

• Case #: IPR2025-00876
• Patent #: 10,218,954
• Filed: May 9, 2025
• Petitioner(s): Google LLC
• Patent Owner(s): Cellular South, Inc.
• Challenged Claims: 1-13

2. Patent Overview

• Title: Video to Data
• Brief Description: The ’954 patent describes a method and system for generating useful data from video content. The system processes video to create separate audio and image files, analyzes them in parallel to extract information like text and objects, and then uses this analysis to generate new, content-rich video information.

3. Grounds for Unpatentability

Ground 1: Claims 1-8 and 11-12 are obvious over Kritt in view of Fontana.

• Prior Art Relied Upon: Kritt (Application # 2014/0181668) and Fontana (Application # 2012/0078712).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that Kritt disclosed the core method of independent claim 1: generating metadata ("video data") from a video by identifying objects and converting audio to text. Kritt described separate "video processing" and "audio processing" modules that could be on different computer systems, implying parallel processing. Fontana was asserted to supplement Kritt by teaching a more detailed implementation of parallel and distributed processing for multimedia content, explicitly describing the concurrent execution of audio and video processing modules and the generation of separate audio files and image files (thumbnails) from a video source for this purpose. Petitioner contended that Kritt’s disclosure of identifying "visual objects" and generating an "audio transcript" met the limitations for object identification and audio-to-text conversion, and its use of "scenes" and "emotions" as contextual topics satisfied further limitations.
  • Motivation to Combine: A Person of Ordinary Skill in the Art (POSA) would combine Kritt with Fontana to improve Kritt’s system. Petitioner asserted that Kritt disclosed a high-level framework but relied on "any known method" for key steps like object identification and speech-to-text recognition. Fontana provided detailed teachings on these very methods, such as specific object recognition programs and speech-to-text algorithms, within an efficient, distributed processing environment. A POSA would have been motivated to incorporate Fontana’s specific, improved techniques into Kritt’s general system to achieve more efficient and robust video analysis.
  • Expectation of Success: A POSA would have had a reasonable expectation of success because both references addressed the same field of multimedia content processing. The combination involved applying Fontana's well-known processing techniques to implement the functionalities broadly described in Kritt, which itself invited the use of such known methods.

Ground 2: Claims 2-3 and 9-10 are obvious over Kritt in view of Fontana and Lau.

• Prior Art Relied Upon: Kritt (Application # 2014/0181668), Fontana (Application # 2012/0078712), and Lau (Application # 2007/0112630).
• Core Argument for this Ground:
  • Prior Art Mapping: This ground built upon the combination of Kritt and Fontana from Ground 1 to address dependent claims related to natural language processing (NLP) and advertisement generation. Petitioner argued that Lau taught using NLP to determine the context of a video for the purpose of matching advertisements. Specifically, Lau disclosed using keywords and recognition techniques to identify subject matter in rich media content. This was asserted to teach the limitations of claims 2 and 3, which required applying NLP to text to determine context or extract a topic. For claims 9 and 10, which required generating and placing an advertisement based on the analyzed text and video data, Petitioner argued that Lau explicitly taught generating advertisements based on keyword matches and placing them at specific times within the video (e.g., pre/mid/post-roll).
  • Motivation to Combine: The motivation was to add a commercially relevant feature (targeted advertising) to the content analysis system of Kritt and Fontana. While Fontana mentioned that advertisement data could be linked to multimedia content, it did not teach how to match the ads to the content. Lau provided this missing element by describing a system for matching ads based on recognized content. A POSA would combine Lau's ad-matching method with the Kritt/Fontana analysis engine to create a complete system that could both analyze video content and monetize it through targeted advertising.
  • Expectation of Success: Success was expected because the combination represented a logical integration of complementary technologies. The text, keywords, and metadata generated by the Kritt/Fontana system would serve as direct inputs for the ad-matching and placement system described by Lau.

Ground 3: Claim 13 is obvious over Kritt.

• Prior Art Relied Upon: Kritt (Application # 2014/0181668).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that claim 13, a system claim, was obvious over Kritt alone because it merely recited a system for performing the method steps of claim 1, which were already disclosed by Kritt. Kritt explicitly described its invention being implemented in a "computer system" with "one or more processors" and "computer readable medium(s)" containing program code. Petitioner contended that these components, as described in Kritt, were configured to execute the exact steps of extracting data from a video that were mapped to Kritt’s teachings in Ground 1, such as converting audio to text and identifying objects in images. Therefore, the system of claim 13 was inherently disclosed or rendered obvious by Kritt’s disclosure of its method.

4. Arguments Regarding Discretionary Denial

• Petitioner argued against discretionary denial under 35 U.S.C. §314(a), asserting that the Fintiv factors weighed against denial. The co-pending district court litigation was in a very early stage, having been recently transferred, with no answer filed, no substantive discovery conducted, and no trial date scheduled.
• Petitioner also argued that denial under 35 U.S.C. §325(d) was unwarranted because the petition raised new invalidity grounds not considered during prosecution. The primary combination of Kritt and Fontana was never before the Examiner. While Lau was mentioned by the Examiner, it was not applied in a rejection or considered in combination with Kritt and Fontana.

5. Relief Requested

• Petitioner requested institution of an inter partes review and cancellation of claims 1-13 of Patent 10,218,954 as unpatentable.