InterDigital Inc v. Hisense Co Ltd

I. Executive Summary and Procedural Information

• Parties & Counsel:
  • Plaintiff: InterDigital, Inc., InterDigital VC Holdings, Inc., and InterDigital Madison Patent Holdings SAS (Delaware / France)
  • Defendant: Hisense Co., Ltd., Hisense USA Corporation, and Hisense Electronics Manufacturing Company of America Corporation (P.R. China / Georgia)
  • Plaintiff’s Counsel: Alston & Bird LLP
• Case Identification: 1:26-cv-00754, N.D. Ga., 02/09/2026
• Venue Allegations: Venue is alleged to be proper in the Northern District of Georgia because two defendant entities, Hisense USA Corporation and Hisense Electronics Manufacturing Company of America Corporation, maintain a regular and established place of business in the district. For the foreign defendant, Hisense Co., Ltd., venue is asserted as proper in any judicial district.
• Core Dispute: Plaintiff alleges that Defendant’s consumer electronics capable of decoding AOMedia Video 1 ("AV1") and displaying High Dynamic Range ("HDR") content infringe six U.S. patents related to video compression and color display technologies.
• Technical Context: The dispute centers on foundational technologies for modern digital video, including advanced video codecs (AV1) that enable efficient streaming and HDR formats (HDR10, HDR10+, Dolby Vision) that provide enhanced picture quality on modern displays.
• Key Procedural History: The complaint does not mention any prior litigation between the parties, Inter Partes Review (IPR) proceedings concerning the asserted patents, or any prior licensing negotiations.

Case Timeline

II. Technology and Patent(s)-in-Suit Analysis

U.S. Patent No. 8,085,846 - "Method and Apparatus for Decoding Hybrid Intra-Inter Coded Blocks"

Issued December 27, 2011 Compl. ¶32

The Invention Explained

• Problem Addressed: Conventional video compression predicted a block of an image using information from either the same image ("intra prediction") or a different image in the sequence ("inter prediction"), but not both simultaneously for the same block Compl. ¶35 This limited the efficiency and quality of the video decoding process Compl. ¶37
• The Patented Solution: The invention discloses a "hybrid intra-inter bi-predictive" decoding method where both intra-frame and inter-frame predictions are combined to create a final, more accurate prediction for a given block or macroblock (’846 Patent, col. 8:2-5; Compl. ¶¶35-36). This hybrid approach is described as an unconventional solution that improves upon methods that use only one predictive mode at a time Compl. ¶39
• Technical Importance: This combined predictive approach is asserted to allow for improved compression efficiency, enhanced video quality, and a reduction in the propagation of data errors during decoding Compl. ¶35 Compl. ¶37

Key Claims at a Glance

• The complaint asserts at least claims 1, 4, 5, 9-11, and 13-22 Compl. ¶77 Independent claim 1 is representative.
• Essential elements of independent claim 1 include:
  • combining a first prediction of a current block with a second prediction of a current block
  • wherein the first prediction of the current block is intra prediction
  • and the second prediction of the current block is inter prediction
• The complaint reserves the right to assert additional claims Compl. ¶83

U.S. Patent No. 9,294,784 - "Method and Apparatus for Region-Based Filter Parameter Selection for De-Artifact Filtering"

Issued March 22, 2016 Compl. ¶40

The Invention Explained

• Problem Addressed: The patent’s background describes issues with prior "de-artifacting" filters used in video decoding. These filters often did not "take the video content itself into account," could not reduce artifacts appearing inside a block, and were not well-suited for complex image areas like edges or textures (’784 Patent, col. 1:40-2:20; Compl. ¶44). Furthermore, signaling filter parameters in the bitstream created undesirable data overhead ’784 Patent, col. 9:1-3 Compl. ¶44
• The Patented Solution: The patent proposes a de-artifact filter that adapts its parameters based on the local statistics of the current block, such as "local variance" ’784 Patent, col. 9:8-12 Compl. ¶¶44-45 Crucially, this "region-based filter parameter is not signaled in the bit stream," which reduces data overhead Compl. ¶45
• Technical Importance: The invention aims to improve decoding performance by reducing compression artifacts more effectively by adapting the filter to the specific video content of a region, without increasing the size of the video bitstream Compl. ¶45

Key Claims at a Glance

• The complaint asserts at least claims 9, 13, and 15 Compl. ¶86 These claims depend from independent claim 1.
• Essential elements of independent claim 1 include:
  • decoding picture data for at least a region of a picture
  • performing de-artifact filtering on the region using a region-based filter parameter
  • wherein the de-artifact filtering is based on local variance of the region
  • and the region-based filter parameter is not signaled in a bit stream
• The complaint reserves the right to present alternative explanations of infringement Compl. ¶92

U.S. Patent No. 10,250,877 - "Method and Device for Coding an Image Block, Corresponding Decoding Method and Decoding Device"

Issued April 2, 2019 Compl. ¶46

• Technology Synopsis: The patent addresses the memory-intensive requirements of prior video compression methods that used adaptive resolution Compl. ¶49 Prior art methods required storing multiple resized versions of reference frames, increasing memory usage (’877 Patent, col. 2:8-13; Compl. ¶51). The invention discloses a method where interpolation and resampling filters are applied jointly, which eliminates the need to store resampled versions of reference images in the decoded picture buffer, thereby reducing memory requirements ’877 Patent, col. 12:1-3 Compl. ¶¶49, 52
• Asserted Claims: At least claims 1 and 4 are asserted Compl. ¶95
• Accused Features: The complaint alleges infringement occurs by "enabling AV1 decoding on the AV1 Accused Instrumentalities" Compl. ¶95

U.S. Patent No. 11,695,962 - "Encoding and Decoding Methods and Corresponding Devices"

Issued July 4, 2023 Compl. ¶54

• Technology Synopsis: The patent is directed to improving the efficiency of entropy coding in video compression, which was limited by systems using fixed block shapes and context models Compl. ¶58 The invention determines a context for decoding a syntax element based on the specific shape of a sub-block (resulting from various "split modes") and a "local template" of neighboring transform coefficients (’962 Patent, col. 18:2-9; Compl. ¶57). By adapting the context model to the data's structure, the method aims to achieve higher compression efficiency and improved visual quality Compl. ¶58
• Asserted Claims: At least claims 1 and 13 are asserted Compl. ¶104
• Accused Features: Infringement is alleged to occur by "enabling AV1 decoding on the AV1 Accused Instrumentalities" Compl. ¶104

U.S. Patent No. 11,399,168 - "Method for Encoding and Method for Decoding a Color Transform and Corresponding Devices"

Issued July 26, 2022 Compl. ¶59

• Technology Synopsis: The patent addresses the difficulty of accurately rendering video across a wide range of display devices with different color capabilities Compl. ¶62 The invention discloses encoding and decoding "target parameters" that describe a target color space, along with "color transform parameters" representing "at least three successively applied color transforms" (’168 Patent, col. 14:9-15; Compl. ¶64). This information, which can be encoded in Supplemental Enhancement Information ("SEI") messages, facilitates the remapping of a video picture from its original color space to the target color space of a specific display device ’168 Patent, col. 15:35-39 Compl. ¶¶64-65
• Asserted Claims: At least claim 18 is asserted Compl. ¶113
• Accused Features: The accused functionality is "enabling HDR playback on the HDR Accused Instrumentalities," defined as devices capable of playing HDR10, HDR10+, or Dolby Vision content Compl. ¶113 Compl. ¶112, n.5

U.S. Patent No. 9,654,751 - "Method, Apparatus and System for Providing Color Grading for Displays"

Issued May 16, 2017 Compl. ¶67

• Technology Synopsis: The patent seeks to solve the problem of consistently displaying a filmmaker's artistic intent across consumer displays with widely varying characteristics like brightness and contrast ratio Compl. ¶70 The invention introduces a system using "virtual device models," which contain parameters specifying display features (’751 Patent, col. 8:65-9:2; Compl. ¶¶70, 72). A consumer display device selects the most equivalent virtual device model by comparing its own specifications to the available models and then adapts the picture data accordingly, thereby rendering a video that more accurately reflects the original color correction ’751 Patent, col. 10:13-44 Compl. ¶73
• Asserted Claims: At least claim 14 is asserted Compl. ¶122
• Accused Features: The accused functionality is "enabling HDR playback on the HDR Accused Instrumentalities" Compl. ¶122

III. The Accused Instrumentality

Product Identification

• The complaint identifies two categories of accused products: "AV1 Accused Instrumentalities" and "HDR Accused Instrumentalities" Compl. ¶6, n.1 Compl. ¶3, n.1
  • AV1 Accused Instrumentalities are defined as "all Hisense devices capable of decoding AV1 content" Compl. ¶3, n.1 Compl. ¶76, n.4
  • HDR Accused Instrumentalities are defined as "all Hisense devices capable of playing back HDR10, HDR10+, or Dolby Vision content" Compl. ¶3, n.1 Compl. ¶112, n.5

Functionality and Market Context

• The complaint alleges that the AV1 Accused Instrumentalities perform video decoding that complies with the AV1 video coding standard, pointing to Defendant's advertisements for televisions that offer applications like Netflix and YouTube, which can playback AV1 content Compl. ¶78 A product page screenshot shows a Hisense Google TV with various streaming application icons, which the complaint uses as evidence of the device's AV1 capabilities Compl. p. 35
• The HDR Accused Instrumentalities are alleged to implement functionality for playing back various HDR formats Compl. ¶113 The complaint includes marketing images and product specifications listing support for "Dolby Vision, HDR 10+, HDR 10" as evidence of this functionality Compl. p. 47 Compl. p. 50
• The complaint asserts that Hisense derives revenue from these activities and that its infringement is "persistent and pervasive," suggesting the products hold commercial significance Compl. ¶1 Compl. ¶76

IV. Analysis of Infringement Allegations

The complaint references exhibits (Exhibits 7-12) that purportedly illustrate infringement but does not attach them (Compl. ¶¶77, 86, 95, 104, 113, 122). The narrative allegations do not contain sufficient detail to construct element-by-element claim charts. The infringement theories are summarized below in prose.

’846 Patent Infringement Allegations

The complaint alleges that Hisense devices infringe the ’846 Patent by "enabling AV1 decoding" Compl. ¶77 The underlying theory appears to be that the AV1 standard, as implemented in the accused devices, necessarily practices the claimed method of combining an intra-frame prediction and an inter-frame prediction to decode a block of video. The complaint does not specify which particular tools or processes within the AV1 decoding pipeline are alleged to perform this combination.

’784 Patent Infringement Allegations

The complaint asserts that the AV1 Accused Instrumentalities infringe the ’784 Patent, again through the act of "enabling AV1 decoding" Compl. ¶86 The theory suggests that the in-loop de-artifacting filters required by the AV1 standard (such as CDEF or the Loop Restoration filter) operate based on local picture characteristics ("local variance") and use parameters that are derived at the decoder rather than being explicitly signaled in the bitstream for each region, thereby meeting the limitations of the asserted claims.

Identified Points of Contention

• Scope Questions: For the ’846 Patent, a central dispute may be whether the prediction methods used in the AV1 standard constitute the "combining" of an "intra prediction" with an "inter prediction" for a single block as required by the claim language. For the ’784 Patent, a key question may be whether the term "local variance" can be construed to read on the specific statistical analyses (e.g., directional pattern detection) performed by AV1’s in-loop filters.
• Technical Questions: A factual question for the ’784 Patent infringement analysis will be whether the parameters for AV1's de-artifacting filters are truly "not signaled in the bit stream" as claimed. The analysis may explore whether high-level control flags or syntax elements in the bitstream indirectly constitute signaling of the filter parameters, which could raise questions of non-infringement.

V. Key Claim Terms for Construction

The Term: "combining a first prediction ... with a second prediction"

(from claim 1 of the ’846 Patent)

Context and Importance

The definition of "combining" is critical, as it forms the central act of the asserted method. The dispute may turn on whether this term requires a specific mathematical operation (e.g., averaging or weighted summing) of one distinct intra prediction and one distinct inter prediction, or if it can be read more broadly to cover any process where both prediction types contribute to the final decoded block.

Intrinsic Evidence for Interpretation

• Evidence for a Broader Interpretation: The specification states that "two (or more) predictions, which may include one or more intra predictions, are to be used (combined) for making the final prediction" ’846 Patent, col. 8:2-5 The parenthetical use of "(combined)" following "to be used" may suggest the terms are interchangeable, supporting a broader meaning.
• Evidence for a Narrower Interpretation: The patent's description of a "hybrid intra-inter bi-predictive" mode may be argued to imply a specific combination of two distinct prediction types, one of each kind, as opposed to a more general use. Specific embodiments or figures illustrating a direct mathematical combination could be used to argue for a narrower construction.

The Term: "based on local variance of the region"

(from claim 1 of the ’784 Patent)

Context and Importance

This term defines the technical basis for the adaptive filtering process. The infringement analysis will depend on whether the methods used by AV1's in-loop filters, which may involve analyzing directional gradients or other local patterns, fall within the scope of being "based on local variance."

Intrinsic Evidence for Interpretation

• Evidence for a Broader Interpretation: The specification states that the filter parameter "adapts the parameter based on the statistics of the current block, such as the local variance" ’784 Patent, col. 9:8-9 The phrase "such as" suggests that "local variance" is an exemplary, rather than exclusive, type of block statistic, potentially supporting a construction that covers other local statistical measures.
• Evidence for a Narrower Interpretation: A defendant may argue that "variance" has a specific, well-understood mathematical meaning in the field of signal processing. If the specification provides a precise formula for calculating "local variance" in an embodiment, it could support an argument that the term should be limited to that specific calculation or its equivalents.

VI. Other Allegations

Indirect Infringement

The complaint alleges inducement of infringement for all asserted patents. The factual basis is Defendant’s alleged encouragement of infringing acts by marketing, selling, and providing instructions for the accused products (Compl. ¶¶80, 89, 98, 107, 116, 125). Specific examples include advertising devices with applications like Netflix and YouTube that use AV1 decoding and promoting the HDR playback capabilities of its products Compl. ¶78 Compl. ¶114

Willful Infringement

For each asserted patent, the complaint alleges that Defendant "has been aware of the [asserted] Patent and its infringement thereof at least as of the filing of this Complaint" (Compl. ¶¶79, 88, 97, 106, 115, 124). This allegation appears to support a claim for post-suit willful infringement rather than pre-suit willfulness.

VII. Analyst’s Conclusion: Key Questions for the Case

This case presents several central questions for the court's determination, rooted in the intersection of patent claim language and the implementation of complex, standardized technologies.

• A primary issue will be one of claim scope versus standard implementation: Do the specific methods for video prediction and filtering as defined in the final AV1 standard fall within the scope of the asserted claims, some of which predate the standard? For the ’846 Patent, this centers on whether AV1's predictive mechanisms constitute the claimed "hybrid intra-inter" combination for a single block.
• A key evidentiary question will concern technical interpretation: For the ’784 Patent, does the AV1 standard's method of adapting in-loop filters based on directional analysis and pattern matching qualify as being "based on local variance" as the claim requires, or is there a fundamental mismatch in the underlying technical operation?
• Finally, the case will likely involve a detailed mapping of product functionality to claim elements: For the HDR-related patents ('168 and '751), the court will need to examine the precise manner in which Hisense devices implement HDR standards like HDR10+ and Dolby Vision to determine if those implementations practice the claimed methods for color space transformation and display adaptation using "virtual device models."