Amazon.com Inc v. Realtime Adaptive Streaming LLC

1. Case Identification

• Case #: IPR2018-01227
• Patent #: 9,578,298
• Filed: June 15, 2018
• Petitioner(s): Amazon.com, Inc., Hulu, LLC, and Netflix, Inc.
• Patent Owner(s): Realtime Adaptive Streaming LLC
• Challenged Claims: 1-19

2. Patent Overview

• Title: Method and Device for Processing a Video Stream of Digital Images
• Brief Description: The ’298 patent describes systems and methods for extracting a 2D-compatible video signal from a stereoscopic (3D) digital video stream. The method involves decoding only one of a pair of stereoscopic images that have been packed into a single "composite frame," allowing the content to be viewed on conventional 2D displays.

3. Grounds for Unpatentability

Ground 1: Anticipation and Obviousness over Tian - Claims 1-13 and 15-19 are anticipated by, or in the alternative obvious over, Tian.

• Prior Art Relied Upon: Tian (Patent 9,036,714).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that Tian taught every limitation of the challenged claims. Tian described methods for processing a video stream where multiple pictures, such as a stereoscopic pair, are combined into a single picture or "frame" using known packing techniques like side-by-side interleaving. This single picture corresponds to the ’298 patent’s "composite frame." Tian further disclosed using metadata, specifically Supplemental Enhancement Information (SEI) messages, to convey the frame-packing information, including the geometry and location of the constituent pictures. Critically, Petitioner asserted Tian explicitly taught "decoding the video picture to provide a decoded representation of at least one of the multiple pictures," directly meeting the key claim limitation of decoding only the part of the composite frame containing the single image needed for 2D display.
  • Motivation to Combine (for §103 grounds): For the alternative obviousness ground, Petitioner argued that to the extent any minor limitation was not explicitly disclosed in a single embodiment, a person of ordinary skill in the art (POSITA) would have found it obvious to combine the various teachings within Tian. For instance, a POSITA would have naturally applied Tian's general disclosure of selective decoding to its specific examples of frame-packed stereoscopic video to achieve the claimed method.
  • Expectation of Success (for §103 grounds): A POSITA would have had a high expectation of success as Tian provided all the necessary components and teachings for selective decoding of frame-packed video.

Ground 2: Obviousness over Lei and the 2010 H.264 Amendment - Claims 1-13 and 15-19 are obvious over Lei in view of the 2010 H.264 Amendment.

• Prior Art Relied Upon: Lei (Application # 2005/0084006) and the 2010 H.264 Amendment (JVT-AE204, a draft standard).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner contended that Lei taught the foundational concepts of the ’298 patent. Lei disclosed a system for handling 3D video where stereoscopic views are combined into a single frame (e.g., as interlaced fields) and taught that it would be "advantageous if only one view...could be decoded to permit viewing on legacy 2D displays." Lei’s system used SEI messages to signal 3D content and triggered a decoder to analyze display capabilities, electing to "decode only one of the current frame interlaced fields" if a non-3D display was detected. While Lei taught the concept, the 2010 H.264 Amendment provided the specific, standardized syntax for frame-packing SEI messages (e.g., frame_packing_arrangement_type) that explicitly defined various packing geometries (side-by-side, top-bottom) and identified the left/right views.
  • Motivation to Combine (for §103 grounds): Petitioner argued a POSITA would combine Lei with the 2010 H.264 Amendment to improve Lei's system using a well-known industry standard. It represented a simple substitution of the specific, standardized syntax from the H.264 Amendment for the more general signaling method in Lei to ensure broader compatibility, achieve better compression, and obtain predictable results.
  • Expectation of Success (for §103 grounds): Success was expected because implementing a specific industry standard (H.264 Amendment) into a system designed for that standard (Lei explicitly mentions H.264) is a routine engineering task.

Ground 3: Obviousness over Tian and Shimada - Claim 14 is obvious over Tian in view of Shimada.

• Prior Art Relied Upon: Tian (Patent 9,036,714) and Shimada (Patent 4,866,542).
• Core Argument for this Ground:
  • Prior Art Mapping: This ground focused on claim 14, which adds the limitation that the decoding method is activated by pressing a "specific remote control button." Petitioner argued that Tian already disclosed activating the selective decoding process based on a "user's command" or "user input." Shimada taught a popular and conventional implementation of user input for electronic devices: a remote-controlling commander with a "push button array" for manual operation.
  • Motivation to Combine (for §103 grounds): Petitioner argued a POSITA would have been motivated to implement Tian's generic "user input" with the specific remote control taught by Shimada for the obvious and well-known benefit of user convenience. Shimada itself taught that remote controls had become a "popular convenience for the users." This combination was presented as a predictable design choice to improve the usability of Tian's system.
  • Expectation of Success (for §103 grounds): A POSITA would have had a very high expectation of success in combining a remote control with a video processing device, as this was a common and well-understood practice in the art.
• Additional Grounds: Petitioner asserted an additional obviousness challenge against claim 14 based on the combination of Lei, the 2010 H.264 Amendment, and Shimada, relying on a similar motivation to add a convenient remote control to the base system.

4. Key Claim Construction Positions

• "composite frame" (Claims 1–3, 6–12): Petitioner proposed this term be construed to mean "a frame containing a pair of stereoscopic images." This construction was based on consistent descriptions in the ’298 patent and was important for mapping the term to prior art disclosures of a "single picture" containing two views.
• "undivided" (Claim 2): Petitioner proposed this term be construed to mean "wholly contained within a determined area of the composite frame." This term was added during prosecution to distinguish over prior art. Petitioner argued that common packing formats like side-by-side, where each image occupies a contiguous half of the frame, met this limitation.

5. Relief Requested

• Petitioner requests institution of an inter partes review and cancellation of claims 1-19 of the ’298 patent as unpatentable.