Alderian Tech LLC v. Lenovo Group Ltd

I. Executive Summary and Procedural Information

• Parties & Counsel:
  • Plaintiff: Alderian Technologies LLC (Texas)
  • Defendant: Lenovo Group Limited, Lenovo Information Products (Shenzhen) Co. Ltd., LCFC (Hefei) Electronics Technology Co. Ltd., Motorola (Wuhan) Mobility Technologies Communication Company Limited, and Lenovo PC HK Ltd. (collectively "Lenovo") (China/Hong Kong)
  • Plaintiff's Counsel: Rubino IP
• Case Identification: 2:26-cv-00428, E.D. Tex., 05/26/2026
• Venue Allegations: The complaint alleges venue is proper because the Defendants do not reside in the United States and may therefore be sued in any judicial district. It further establishes Defendants' contacts with the district through sales, marketing, distribution channels, and established places of business within Texas.
• Core Dispute: Plaintiff alleges that Defendant's consumer electronic devices, including smartphones, tablets, and laptops, infringe two patents related to video compression and coding methodologies.
• Technical Context: The technology involves digital video compression, a foundational process for enabling the efficient storage and transmission of video data across a wide range of modern electronic devices.
• Key Procedural History: The complaint notes that courts within the Eastern District of Texas have previously found personal jurisdiction over Lenovo Group Ltd. in other patent cases, and that the Federal Circuit denied Lenovo's request for a writ of mandamus to challenge those findings.

Case Timeline

Date	Event
2005-09-15	Earliest Priority Date for '172 Patent
2010-06-07	Motorola, Inc. allegedly cited the '172 Patent family in its own patent application
2015-06-16	'172 Patent Issued
2018-10-26	Earliest Priority Date for '387 Patent
2020-05-05	'387 Patent Issued
2026-05-26	Complaint Filed

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

U.S. Patent No. 10,645,387 - "Predictive Quantization Coding Method and Video Compression System"

• Patent Identification: U.S. Patent No. 10,645,387, "Predictive Quantization Coding Method and Video Compression System," issued May 5, 2020 Compl. ¶33

The Invention Explained

• Problem Addressed: The patent describes shortcomings in existing video compression techniques, including the misjudgment of prediction pixels, underutilization of correlation between pixel textures, and high computational complexity, all of which limit the achievable data compression ratio and can increase distortion loss '387 Patent, col. 1:36-44
• The Patented Solution: The invention proposes a method for predictive coding that involves analyzing "texture direction gradients" of a pixel's components (e.g., R, G, B) to determine reference pixels for prediction '387 Patent, abstract After calculating a "prediction residual" (the difference between the actual and predicted pixel value), the method employs a multi-stage process involving quantization and two distinct "rate distortion optimizations" to generate a more efficient compressed data stream '387 Patent, col. 2:1-6 '387 Patent, col. 7:55-59
• Technical Importance: The described method aims to improve video compression efficiency, thereby reducing the bandwidth required for transmission, while also minimizing distortion and complexity '387 Patent, col. 1:45-49

Key Claims at a Glance

• The complaint asserts at least independent Claim 1 Compl. ¶43
• The essential elements of Claim 1 include:
  • (a) dividing a pixel to be processed into a plurality of pixel components.
  • (b) obtaining one pixel component to be processed.
  • (c) obtaining texture direction gradients of the pixel component.
  • (d) obtaining reference pixels according to the texture direction gradients and positional relationships.
  • (e) obtaining a prediction residual of the pixel component.
  • (f) repeating the process for each pixel component to form a prediction residual code stream.
  • (g) dividing the prediction residual code stream into a plurality of quantization units.
  • (h) obtaining first and second rate distortion optimizations to create a final quantization residual code stream.
• The complaint does not explicitly reserve the right to assert dependent claims.

U.S. Patent No. 9,060,172 - "Methods and Systems for Mixed Spatial Resolution Video Compression"

• Patent Identification: U.S. Patent No. 9,060,172, "Methods and Systems for Mixed Spatial Resolution Video Compression," issued June 16, 2015 Compl. ¶34

The Invention Explained

• Problem Addressed: The patent addresses the general need for video compression to reduce the large quantity of data required to represent video content, which is necessary for efficient storage and transmission '172 Patent, col. 1:20-25
• The Patented Solution: The invention describes a "mixed spatial resolution" encoding system. For a given video frame, the system generates two separate compressed versions: one at the original full resolution and another at a reduced spatial resolution, which is then upsampled back to the original size '172 Patent, abstract The system then compares a characteristic (e.g., distortion) of both reconstructed frames against the original frame and selects for output the coded data corresponding to the version that has the "lesser deviation" '172 Patent, col. 2:2-25 '172 Patent, FIG. 1
• Technical Importance: This approach provides a method to dynamically improve compression efficiency by selectively reducing the spatial resolution of certain frames when doing so does not perceptibly degrade visual quality '172 Patent, col. 1:62-66

Key Claims at a Glance

• The complaint asserts at least independent Claim 1 Compl. ¶60
• The essential elements of Claim 1 include:
  • generating a reconstructed full resolution frame and corresponding coded data from an original full resolution frame.
  • generating an upsampled reconstructed reduced spatial resolution frame and corresponding coded data from the original full resolution frame.
  • determining a first deviation by comparing the reconstructed full resolution frame to the original.
  • determining a second deviation by comparing the upsampled reconstructed reduced spatial resolution frame to the original.
  • outputting the coded data for the full resolution frame if its deviation is less than the reduced resolution frame's deviation, and otherwise outputting the coded data for the reduced resolution frame.
• The complaint does not explicitly reserve the right to assert dependent claims.

III. The Accused Instrumentality

Product Identification

• The accused products are a broad category of Lenovo devices, including smartphones, tablets, laptops, workstations, and desktops Compl. ¶¶37-38 The complaint identifies the "Lenovo Tab Extreme," which is equipped with a MediaTek Dimensity 9000 System-on-Chip (SoC), as an exemplary infringing product Compl. ¶42 Compl. ¶59

Functionality and Market Context

• The complaint alleges that the accused products contain SoCs, chipsets, and software that utilize hardware and software encoders and decoders to perform video compression Compl. ¶¶37-38 The infringement allegations center on the functionality of these video codecs, which are alleged to implement the patented methods Compl. ¶42 Compl. ¶59 The complaint provides a screenshot of a directory structure for a MediaTek video codec ("mtk-vcodec") from an Android open-source repository as evidence of the accused functionality Compl. ¶46, p. 23 Compl. ¶63, p. 29 This screenshot shows separate directories for video encoding ("venc") and decoding ("vdec"), indicating the presence of the accused codec components.

IV. Analysis of Infringement Allegations

'387 Patent Infringement Allegations

Claim Element (from Independent Claim 1)	Alleged Infringing Functionality	Complaint Citation	Patent Citation
(a) dividing a pixel to be processed into a plurality of pixel components...	The Lenovo Tab Extreme performs a quantization coding method which involves dividing a pixel to be processed into a plurality of pixel components.	¶44(a)	col. 2:7-11
(c) obtaining texture direction gradients of the pixel component to be processed	The method includes obtaining texture direction gradients of the pixel component to be processed.	¶44(c)	col. 5:1-11
(d) obtaining reference pixels according to the texture direction gradients and positional relationships between the pixel component to be processed and A remaining of the plurality of pixel components	The method includes obtaining reference pixels according to texture direction gradients and positional relationships.	¶44(d)	col. 5:44-49
(f) ...forming a prediction residual code stream including the prediction residuals of the pixel to be processed of the image	The method includes repeating steps to form a prediction residual code stream including the prediction residuals of the pixel.	¶44(f)	col. 7:41-49
(h) obtaining first rate distortion optimizations and second rate distortion optimizations corresponding to the plurality of quantization units to obtain a quantization residual code stream	The method includes obtaining first and second rate distortion optimizations to obtain a quantization residual code stream. The complaint alleges this is performed by HEVC and other codecs using inter-/intra-picture prediction.	¶44(h)	col. 7:55-59

• Identified Points of Contention:
  • Scope Questions: A central question may be whether the term "texture direction gradients" as defined and used in the patent can be read to cover the standardized intra-prediction modes (e.g., angular prediction) found in codecs like HEVC, which the complaint alleges are used in the accused products Compl. ¶44 The defense may argue that the patent describes a specific, proprietary calculation for these gradients that is technically distinct from the methods used in industry-standard codecs.
  • Technical Questions: The infringement analysis may turn on whether the accused products' HEVC codecs perform the specific two-stage "rate distortion optimization" process required by claim 1(h). The complaint alleges this occurs Compl. ¶44(h), but the court will likely require detailed evidence comparing the operation of the accused codecs to the patent's specific two-step compensation and optimization method '387 Patent, col. 9:42-56

'172 Patent Infringement Allegations

Claim Element (from Independent Claim 1)	Alleged Infringing Functionality	Complaint Citation	Patent Citation
generating a reconstructed full resolution frame and coded data corresponding to the reconstructed full resolution frame from an original full resolution frame	The Lenovo Tab Extreme performs a method that includes generating a reconstructed full resolution frame and its corresponding coded data from an original frame.	¶61	col. 5:51-54
generating an upsampled reconstructed reduced spatial resolution frame and coded data corresponding to the upsampled reconstructed reduced spatial resolution frame from the original full resolution frame	The method includes generating an upsampled reconstructed reduced spatial resolution frame and its corresponding coded data from the original frame.	¶61	col. 5:62-63
determining a first deviation by comparing at least one characteristic in the reconstructed full resolution frame with said at least one characteristic in the original full resolution frame	The method includes determining a first deviation by comparing a characteristic in the reconstructed full resolution frame with the same characteristic in the original frame.	¶61	col. 5:64-66
determining a second deviation by comparing said at least one characteristic in the upsampled reconstructed reduced spatial reduced resolution frame with said at least one characteristic in the original full resolution frame	The method includes determining a second deviation by comparing the same characteristic in the upsampled reduced resolution frame with that characteristic in the original frame.	¶61	col. 6:1-2
outputting the coded data corresponding to the reconstructed full resolution frame when the first deviation is less than the second deviation; otherwise outputting the coded data...	The method includes outputting the coded data for the frame version (full or reduced resolution) that has the lesser deviation.	¶61	col. 6:2-9

• Identified Points of Contention:
  • Functional Questions: A key point of contention will likely be whether the accused MediaTek SoCs actually perform the claimed method of generating two separate, fully reconstructed frames (one full-res, one upsampled reduced-res) and then selecting between them based on a deviation comparison. The complaint makes a conclusory allegation that mirrors the claim language Compl. ¶61, but it does not provide specific evidence, such as source code or technical documentation, demonstrating this specific dual-path, compare-and-select architecture.
  • Scope Questions: The dispute may involve the scope of "determining a...deviation." The defense could argue that standard rate-control algorithms in modern codecs make encoding decisions based on rate-distortion optimization (RDO) before committing to a final encoding path, which is functionally different from the patent's method of fully encoding and reconstructing two versions and then comparing them after the fact.

V. Key Claim Terms for Construction

• '387 Patent:

  • The Term: "texture direction gradients" (Claim 1(c))
  • Context and Importance: This term is fundamental to the patent's predictive coding step. Its construction will determine whether standardized prediction techniques, such as the angular intra-prediction modes in HEVC, fall within the scope of the claim. Practitioners may focus on this term because the infringement case depends on mapping this specific patent language onto the operation of a widely used industry standard.
  • Intrinsic Evidence for Interpretation:
    • Evidence for a Broader Interpretation: The specification describes the term's purpose as utilizing "texture correlation for predictive coding" '387 Patent, col. 4:23-25, which could support an argument that any method using directional pixel information for prediction is covered.
    • Evidence for a Narrower Interpretation: The specification provides a very specific method for calculating these gradients, which involves defining a "texture reference component" (the "O" pixel in FIG. 3) and forming vectors to surrounding pixels to determine gradient direction and size '387 Patent, col. 5:12-33 '387 Patent, FIG. 4 This could support a narrower construction limited to this specific implementation.

• '172 Patent:

  • The Term: "outputting the coded data corresponding to the reconstructed full resolution frame when the first deviation is less than the second deviation" (Claim 1)
  • Context and Importance: This limitation defines the core decision-making logic of the invention. The entire infringement theory rests on whether the accused products perform this explicit A/B test between two fully processed options.
  • Intrinsic Evidence for Interpretation:
    • Evidence for a Broader Interpretation: The patent's summary describes the invention more generally as identifying "video frames whose spatial resolution can be reduced and then re-interpolated without visible quality degradation" '172 Patent, col. 1:62-65, potentially allowing for other decision-making processes that achieve this goal.
    • Evidence for a Narrower Interpretation: The abstract and the claim language itself describe a specific sequence: generate two versions, determine two deviations, compare the deviations, and then select one for output '172 Patent, abstract The flowchart in FIG. 1 visually reinforces this specific two-path, compare-and-select workflow, which could support a construction requiring this literal implementation '172 Patent, FIG. 1, steps 180a, 180b, 190

VI. Other Allegations

• Indirect Infringement: The complaint alleges inducement of infringement for both patents. The factual basis is that Lenovo provides the accused products to customers and supplies instructional materials-such as user manuals, guides, and website support-that allegedly instruct end-users on how to operate the products in an infringing manner Compl. ¶¶52-53 Compl. ¶¶68-69 A screenshot of Lenovo's support website, showing how users can find product manuals, is provided as evidence of these affirmative acts of inducement Compl. p. 26, fig. 38 Compl. p. 32, fig. 44
• Willful Infringement: The complaint alleges willful infringement based on knowledge of the patents, at a minimum, from the filing of the complaint Compl. ¶51 Compl. ¶67 It also alleges willful blindness, claiming Lenovo has a "policy of not reviewing the patents of others" to avoid knowledge of infringement Compl. ¶51 Compl. ¶67 Critically, for the '172 Patent, the complaint alleges pre-suit knowledge dating back to June 2010, based on the assertion that Motorola, Inc. (now a Lenovo subsidiary) cited the '172 patent family during the prosecution of its own patent applications Compl. ¶67 Compl. ¶67, fn 43

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

1. Definitional Scope: A core issue for the '387 Patent will be one of definitional scope: can the term "texture direction gradients", which is described with a specific calculation method in the patent, be construed broadly enough to encompass the standardized angular intra-prediction techniques used in the accused HEVC codecs, or is it limited to the patent's novel implementation?
2. Evidentiary Sufficiency: A key evidentiary question for the '172 Patent will be one of functional operation: does the Plaintiff have sufficient evidence to prove that Lenovo's products perform the specific claimed method of fully encoding a frame down two parallel paths (full and reduced resolution), comparing the results, and selecting one for output, or do the accused products use a different, non-infringing rate-control algorithm?
3. Pre-Suit Knowledge and Willfulness: The allegation that Motorola (a Lenovo entity) cited the '172 patent family in 2010 raises a significant factual question regarding pre-suit knowledge. The resolution of this issue will be critical to the claim for willful infringement of the '172 Patent, potentially exposing Lenovo to enhanced damages.