Arashi Vision Inc v. GoPro Inc

1. Case Identification

• Case #: IPR2024-01052
• Patent #: 11,336,832
• Filed: June 20, 2024
• Petitioner(s): ARASHI VISION (U.S.) LLC (D/B/A INSTA360)
• Patent Owner(s): GoPro, Inc.
• Challenged Claims: 1-20

2. Patent Overview

• Title: Horizon Leveling Videos
• Brief Description: The ’832 patent discloses systems and methods for generating horizon-leveled video content from a video captured by a potentially tilted or rotated image capture device. The technology involves determining a "viewing window" within the captured visual content and using its placement, size, and rotation to compensate for the device's motion, thereby creating a stabilized output video.

3. Grounds for Unpatentability

Ground 1: Obviousness over Thomason and Pacurariu - Claims 1, 2, 7-12, and 17-20 are obvious over Thomason in view of Pacurariu.

• Prior Art Relied Upon: Thomason (Application # 2005/0168583) and Pacurariu (Application # 2015/0181123).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that Thomason taught the core concept of horizon leveling by correcting for camera orientation. Thomason disclosed applying a rotation correction to a captured image and then cropping it to eliminate blank space, which Petitioner equated to the claimed "viewing window." Pacurariu was argued to teach storing tilt angle data for each video frame as metadata. The combination, Petitioner asserted, disclosed a system that obtains rotational position information for each frame of a video (Pacurariu) and uses that information to determine a viewing window (Thomason’s cropping) to generate horizon-leveled content, thus meeting the limitations of independent claims 1 and 11. Petitioner further mapped Thomason’s disclosure of a 1:1 aspect ratio to claim 2 and its description of cropping based on rotation angle to claim 7’s size determination.
  • Motivation to Combine: Petitioner contended a POSITA would combine these references as both address the same technical problem of improving video content by compensating for camera motion. A POSITA would have been motivated to integrate Pacurariu’s method of associating tilt data with each video frame into Thomason’s system to enable more precise, frame-by-frame rotation correction.
  • Expectation of Success: Combining these known techniques for rotation correction and data management would predictably yield a more effective and efficient horizon-leveling system.

Ground 2: Obviousness over Thomason, Pacurariu, and Watanabe - Claims 3-6 and 13-16 are obvious over Thomason and Pacurariu in view of Watanabe.

• Prior Art Relied Upon: Thomason (Application # 2005/0168583), Pacurariu (Application # 2015/0181123), and Watanabe (Patent 10,778,896).
• Core Argument for this Ground:
  • Prior Art Mapping: This ground built upon the Thomason/Pacurariu combination to address claims reciting the inclusion of distortion. Petitioner argued that Watanabe taught image processing techniques to make shake-corrected video appear more natural by intentionally adding distortion. Specifically, Watanabe disclosed applying barrel distortion to a non-distorted image, which Petitioner asserted met the limitations of claims 3 and 13. Watanabe also explicitly disclosed barrel and pincushion distortion (claim 4), distortion from wide-angle lenses (claim 5), and applying distortion to non-distorted content (claim 6).
  • Motivation to Combine: A POSITA, having developed the horizon-leveling system of Thomason and Pacurariu, would be motivated to improve the aesthetic quality of the output video. Petitioner argued that a POSITA would look to known techniques, such as those in Watanabe, to add desirable distortion that makes the corrected video appear more natural and less jarring to the viewer.
  • Expectation of Success: Applying Watanabe’s known distortion-addition techniques to the horizon-leveled video from the primary combination would predictably result in a more visually appealing final product.

Ground 3: Obviousness over Voss and Derbanne - Claims 1, 2, 7-12, and 17-20 are obvious over Voss in view of Derbanne.

• Prior Art Relied Upon: Voss (WO 2015/183636) and Derbanne (Patent 10,341,564).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner asserted that Voss disclosed a system that dynamically maintains a horizontal video framing by transforming an "inset rectangle" (the viewing window) to compensate for device rotation measured by a gyroscope. This taught determining the placement of a viewing window based on rotational positions to generate horizon-leveled content. Derbanne was argued to supplement Voss by teaching a system that captures a larger "optical field of view" and generates stabilized video from smaller "punch-outs" (viewing windows) within that larger view, explicitly disclosing that video content is viewable as a function of progress through a progress length.
  • Motivation to Combine: Petitioner stated that both Voss and Derbanne relate to video stabilization by compensating for device motion and are from the same field of endeavor. A POSITA would be motivated to combine the teachings to improve the video management and processing capabilities of Voss's system, yielding a predictable improvement.
  • Expectation of Success: Combining the specific real-time inset rectangle transformation of Voss with the broader punch-out framework of Derbanne was argued to be a straightforward integration of complementary techniques with a high expectation of success.
• Additional Grounds: Petitioner asserted an additional obviousness challenge against claims 3-6 and 13-16 based on the combination of Voss, Derbanne, and Watanabe, relying on Watanabe to teach the distortion limitations.

4. Arguments Regarding Discretionary Denial

• Petitioner argued that discretionary denial under Fintiv is inappropriate. It contended that the plain language of the Fintiv factors is directed to parallel district court litigation, not proceedings at the U.S. International Trade Commission (ITC), where a related matter is pending. Petitioner noted that ITC invalidity rulings are not binding on the USPTO.
• Petitioner also argued against denial under §325(d). It asserted that none of the primary prior art references relied upon in the petition (Thomason, Pacurariu, Watanabe, Voss, or Derbanne) were substantively considered by the Examiner during the original prosecution of the ’832 patent. Petitioner claimed this failure to consider highly relevant art that was not before the Office constituted a material error.

5. Relief Requested

• Petitioner requests the institution of an inter partes review and the cancellation of claims 1-20 of Patent 11,336,832 as unpatentable.