Samsung Electronics Co Ltd v. Image Processing Technologies LLC

1. Case Identification

• Case #: IPR2017-01212
• Patent #: 8,989,445
• Filed: March 30, 2017
• Petitioner(s): Samsung Electronics Co., Ltd.; and Samsung Electronics America, Inc.
• Patent Owner(s): Image Processing Technologies, LLC
• Challenged Claims: 2-3, 5, 7-8, 10-17, 19-23, 26, and 28-30

2. Patent Overview

• Title: Image Processing Method and Apparatus, and Target Tracking Method and Apparatus
• Brief Description: The ’445 patent describes methods for identifying and tracking a target within a video signal. The core technology involves generating one or more histograms based on pixel characteristics from successive video frames to determine a target's location and movement.

3. Grounds for Unpatentability

Ground 1: Obviousness over Gerhardt and Bassman - Claims 3, 5, 7-8, 14-17, 26, and 28-29 are obvious over Gerhardt in view of Bassman.

• Prior Art Relied Upon: Gerhardt (Patent 5,481,622) and Bassman (Patent 6,044,166).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that the combination of Gerhardt and Bassman taught all limitations of the challenged claims. Gerhardt disclosed an eye-tracking system that identified a user's pupil by generating an intensity histogram from video frames, locating the pupil blob, and mapping its coordinates. Bassman disclosed a vehicle-tracking system that also used intensity histograms from successive video frames to track targets over time. Petitioner asserted that these references collectively taught the core process of receiving a video signal, generating a histogram from a first frame to identify a target, generating a histogram from a subsequent frame, and adjusting the target's location based on the second histogram.
  • Motivation to Combine: A Person of Ordinary Skill in the Art (POSITA) would combine Gerhardt and Bassman because both address the same problem of tracking targets in a video stream using histograms. Petitioner contended a POSITA would have recognized that Bassman’s method of explicitly linking a target's position between frames could improve the reliability of Gerhardt’s system, particularly in overcoming issues like a user blinking, which could cause tracking to fail.
  • Expectation of Success: The combination involved applying known image processing techniques to similar systems, leading to a predictable improvement in tracking reliability and speed.

Ground 2: Obviousness over Gerhardt, Bassman, and Horn - Claims 2, 10, and 11 are obvious over Gerhardt in view of Bassman, and further in view of Horn.

• Prior Art Relied Upon: Gerhardt (Patent 5,481,622), Bassman (Patent 6,044,166), and Horn (a 1981 journal article titled Determining Optical Flow).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that Horn supplied the teachings of smoothing an input signal and determining a target’s speed, as required by claims 2 and 10-11. Horn, a foundational paper on computer vision, described methods for calculating "optical flow" to determine the velocity of brightness patterns between image frames. This process inherently involved calculating target speed (velocity vectors) and using methods that smooth the input signal to reduce noise and enable accurate calculations.
  • Motivation to Combine: A POSITA would combine Horn with the Gerhardt/Bassman system to improve its performance. Since video signals are inherently noisy, a POSITA would have been motivated to apply Horn's well-known smoothing techniques to achieve clearer target identification. Furthermore, to add the functionality of determining target speed (as claimed), a POSITA would logically turn to Horn's seminal work on optical flow for a method to do so.
  • Expectation of Success: Combining Horn’s established algorithms for smoothing and velocity calculation with a standard histogram-based tracking system was a straightforward application of known principles with a high expectation of success.

Ground 3: Obviousness over Gerhardt, Bassman, and Tsuchikawa - Claims 3, 12, and 13 are obvious over Gerhardt in view of Bassman, and further in view of Tsuchikawa.

• Prior Art Relied Upon: Gerhardt (Patent 5,481,622), Bassman (Patent 6,044,166), and Tsuchikawa (Patent 5,748,775).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner asserted that Tsuchikawa taught generating multiple histograms in multiple domains and identifying non-moving areas, as required by claims 12 and 13. Tsuchikawa disclosed an improved method of background subtraction for target extraction, which involved identifying non-moving areas of an image. It explicitly taught creating histograms from various "image feature parameters," including not just intensity but also color (R, G, B), hue, and saturation, thereby teaching the use of multiple histograms across different domains.
  • Motivation to Combine: A POSITA would incorporate Tsuchikawa’s teachings to enhance the stability and accuracy of the Gerhardt/Bassman system. Using histograms from multiple domains (e.g., color) would allow for more robust target differentiation, while improved background subtraction would make tracking more reliable in environments with changing illumination or gradual background movement.
  • Expectation of Success: Adding multi-domain analysis and background subtraction from Tsuchikawa to a base tracking system represented a predictable enhancement to improve tracking robustness.

Ground 4: Obviousness over Gerhardt, Bassman, and Kuwashima - Claims 19-23 and 30 are obvious over Gerhardt in view of Bassman, and further in view of Kuwashima.

• Prior Art Relied Upon: Gerhardt (Patent 5,481,622), Bassman (Patent 6,044,166), and Kuwashima (Patent 5,764,786).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that Kuwashima provided the missing elements related to actuating a servomotor to physically adjust a camera based on the tracked target's location, as required by claims 19-21 and 30. Kuwashima disclosed a system that calculated the coordinates of a moving object and used a "position control servo circuit" to control motors that automatically move a camera head to track the object. It also disclosed tracking a human face as a target.
  • Motivation to Combine: A POSITA would combine Kuwashima’s camera control system with the Gerhardt/Bassman tracking logic to create a fully automated tracking system. This combination represented a logical next step from simply tracking a target on-screen to physically following it with a camera, which would be a desirable feature for applications like security or automated videoconferencing.
  • Expectation of Success: Integrating a known servo-based camera positioning system (Kuwashima) with a known electronic image tracking system (Gerhardt/Bassman) was a well-understood engineering task with a predictable outcome.

4. Arguments Regarding Discretionary Denial

• Petitioner argued that discretionary denial under 35 U.S.C. §325(d) would be inappropriate. It asserted this petition was not redundant to its previously filed IPR (IPR2017-00357) because it was necessitated by the Patent Owner's amendment of infringement contentions in parallel district court litigation to include new claims. Petitioner further contended that this petition raised new arguments based on entirely different prior art references not included in the earlier petition.

5. Relief Requested

• Petitioner requests institution of an inter partes review and cancellation of claims 2-3, 5, 7-8, 10-17, 19-23, 26, and 28-30 of the ’445 patent as unpatentable.