DCT

2:23-cv-00617

Hanshow Technology Co Ltd v. SES imagotag SA

Log In

I. Executive Summary and Procedural Information

  • Parties & Counsel:
  • Case Identification: 2:23-cv-00617, E.D. Tex., 12/19/2023
  • Venue Allegations: Venue is asserted based on Defendants being foreign corporations not resident in the United States, and on allegations that Defendants maintain a regular and established place of business in Irving, Texas.
  • Core Dispute: Plaintiff alleges that Defendant’s VUSION digital shelf platform, which includes electronic shelf labels and camera-based monitoring systems, infringes patents related to automated retail inventory analysis and distributed video processing.
  • Technical Context: The technology at issue involves using imaging systems and artificial intelligence to automate the monitoring of products on retail shelves, a process intended to improve inventory accuracy and operational efficiency.
  • Key Procedural History: The complaint is the initiating document in this litigation. No prior litigation, licensing history, or other procedural events are mentioned.

Case Timeline

Date Event
2015-02-27 U.S. Patent No. 10,701,321 Priority Date
2016-03-29 U.S. Patent No. 11,087,272 Priority Date
2020-06-30 U.S. Patent No. 10,701,321 Issued
2021-08-10 U.S. Patent No. 11,087,272 Issued
2023-12-19 Complaint Filed

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

U.S. Patent No. 11,087,272: "System and method for locating, identifying and counting items" (Issued Aug. 10, 2021)

The Invention Explained

  • Problem Addressed: The patent describes the process of manually monitoring retail inventory as expensive, time-consuming, and insufficient for identifying misplaced or damaged products. It notes that existing machine vision systems often require a detailed, pre-existing "planogram" (a diagram of intended product placement), which itself requires significant human intervention to create and maintain (’272 Patent, col. 1:26-54).
  • The Patented Solution: The invention proposes an automated system, potentially mounted on a robot, that creates an inventory map without an initial planogram. The system captures images of shelves, detects shelf labels, defines "bounding boxes" around products, and associates the label content with the corresponding product images. This process is used to build a training dataset that trains a "product classifier," effectively creating a product library from the ground up (’272 Patent, Abstract; col. 2:1-4). A flowchart included in the complaint illustrates a process for creating a product space without requiring an initial planogram (Compl. ¶18, referencing ’272 Patent, Fig. 6).
  • Technical Importance: This approach aims to solve a key bottleneck in automated retail analytics by removing the dependency on manually created planograms, which could enable more scalable and dynamic inventory tracking.

Key Claims at a Glance

  • The complaint asserts independent claims 1 (method) and 18 (system) (Compl. ¶24).
  • Independent Claim 1 recites a method comprising the key steps of:
    • capturing one or more images of a portion of a shelf with a camera;
    • detecting and reading content of a shelf label;
    • defining one or more product facing bounding boxes;
    • associating the shelf label content with each product facing bounding box;
    • cross-correlating the bounding boxes across one or more stores by matching shelf label content;
    • building training data based on this association and cross-correlation;
    • using the training data to build a product classifier; and
    • using the product classifier to classify an item as the product (Compl. ¶19; ’272 Patent, col. 15:15–col. 16:21).
  • The complaint reserves the right to assert additional claims (Compl. ¶30).

U.S. Patent No. 10,701,321: "System and method for distributed video analysis" (Issued Jun. 30, 2020)

The Invention Explained

  • Problem Addressed: The patent identifies high data traffic volume as a major problem in conventional video surveillance systems, where entire video streams are transmitted over networks for processing. This consumes significant bandwidth and requires powerful, expensive computer systems for analysis (’321 Patent, col. 2:46-67).
  • The Patented Solution: The invention describes a distributed architecture that separates the video processing workload into two distinct functions: a local "object detection system" and a remote "video analytics system." The local system detects objects of interest in the video feed and extracts only the relevant portions of the video frames (e.g., cropped images of the objects). Only these extracted portions are sent over the network to the remote system for detailed analysis, thereby reducing bandwidth consumption (’321 Patent, Abstract; col. 2:8-20). The complaint includes a block diagram from the patent illustrating this distributed system (Compl. ¶21, referencing ’321 Patent, Fig. 1).
  • Technical Importance: By minimizing the amount of data transmitted, this architecture could enable more efficient and cost-effective deployment of large-scale video analytics systems, particularly over networks with limited bandwidth.

Key Claims at a Glance

  • The complaint asserts independent claims 1 (system) and 13 (method) (Compl. ¶32).
  • Independent Claim 1 recites a video surveillance system comprising:
    • one or more surveillance cameras;
    • an object detection system that detects objects of interest, extracts portions of frames containing them, and sends the extracted portions;
    • a video analytics system that receives only the extracted portions from the object detection system over a network and analyzes them;
    • wherein the video analytics system is operated by a different business entity than the cameras; and
    • wherein the object detection system ranks the image data based on factors like proximity and picture quality (Compl. ¶22; ’321 Patent, col. 11:43–col. 12:12).
  • The complaint reserves the right to assert additional claims (Compl. ¶38).

III. The Accused Instrumentality

Product Identification

  • The accused instrumentalities are the "VUSION digital shelf platform," which includes "Captana devices, systems, and/or functionality, and VUSION electronic shelf labels ('VUSION ESLs')" (Compl. ¶9).

Functionality and Market Context

  • The complaint alleges that Defendants market the accused products as an "IoT Cloud technology" platform for retail, referencing components such as the "VUSION Retail IoT Cloud platform, the Captana Sensor Cloud, Pulse Data Analytics and Edge Digital Media" (Compl. Ex. 4 at 10-11).
  • Visual evidence submitted with the complaint suggests the Captana system is marketed for global retail operations, with one graphic indicating it "runs already in more than 250 Stores" and monitors "More than 1. Mio SKUs... daily" (Compl. Ex. 3 at 93). This image illustrates the "Global Expansion" of the accused technology. Another visual highlights Defendants' U.S. presence with "4 offices" including a new one in Dallas, Texas (Compl. Ex. 3 at 155).
  • The complaint alleges these products are used, sold, and offered for sale within the United States and the State of Texas (Compl. ¶15).

IV. Analysis of Infringement Allegations

The complaint alleges infringement of claims from the ’272 and ’321 patents and states that claim comparison charts are attached as Exhibits 7 and 8, respectively (Compl. ¶¶ 24, 32). However, these exhibits were not filed with the complaint. The complaint itself does not provide a narrative mapping of specific features of the accused VUSION platform to the elements of the asserted claims. It makes conclusory allegations that the accused products infringe. Therefore, a detailed claim chart summary cannot be constructed from the provided document.

  • Identified Points of Contention:
    • ’272 Patent: A primary question will be whether the accused VUSION platform performs the specific method recited in claim 1. Key factual disputes may arise over whether the system (1) defines "product facing bounding boxes," (2) "cross-correlates" those boxes "across one or more stores" by matching shelf label content, and (3) uses that cross-correlated data to "build a product classifier" in the manner claimed.
    • ’321 Patent: The central infringement question will be architectural. A court will need to determine if the VUSION platform embodies the specific distributed architecture of claim 1, which requires a functional and, in some respects, organizational separation between an "object detection system" and a "video analytics system." Factual questions will include whether the accused system sends "only the extracted portions of the frames" over the network and whether the analytics system is "operated by a different business entity than the one or more surveillance cameras," as the claim requires.

V. Key Claim Terms for Construction

  • '272 Patent:

    • The Term: "building training data based on the association of the shelf label content with the cross-correlated product facing bounding boxes"
    • Context and Importance: This term is the inventive core of the method, describing how the system learns to identify products without a pre-existing planogram. The construction of this entire phrase will be critical to determining infringement, as it defines the specific mechanism for creating the product classifier.
    • Intrinsic Evidence for Interpretation:
      • Evidence for a Broader Interpretation: The claim language itself does not specify the exact algorithm, which could support an interpretation covering various methods of creating a dataset from associated labels and images.
      • Evidence for a Narrower Interpretation: The specification describes a specific process where image descriptors are extracted from the bounding box and associated with an identifier from the label, which is then used to train classifiers. For example, it states "the image descriptors can be classified and labelled with the identifier" to "automatically build[] a product library" (’272 Patent, col. 12:45-53). This could support a narrower construction requiring these specific sub-steps.

  • ’321 Patent:

    • The Term: "a video analytics system is operated by a different business entity than the one or more surveillance cameras"
    • Context and Importance: This limitation imposes not just a technical but also a business-relationship requirement on the infringing system. Its construction is crucial, as infringement may depend on the corporate structure and operational relationship between the provider of the on-site cameras and the provider of the cloud-based analytics. Practitioners may focus on this term because it creates a potential non-infringement argument based on how the system is deployed and managed for a given customer.
    • Intrinsic Evidence for Interpretation:
      • Evidence for a Broader Interpretation: The term could be interpreted to mean any scenario where the legal entities owning the cameras and the analytics software are distinct, even if they are part of the same overall service offering to an end-customer.
      • Evidence for a Narrower Interpretation: The specification provides examples where a third-party company operates a "cloud service" for multiple, separate business clients like "ACME," "Biz Corp," and "Cam Corp" (’321 Patent, col. 4:38-44, Fig. 1). This could support a narrower construction requiring a multi-tenant, third-party service provider model, as opposed to a single vendor providing an integrated hardware and software solution to one customer.

VI. Other Allegations

  • Indirect Infringement: The complaint alleges both induced and contributory infringement. The inducement allegations are based on Defendants allegedly providing "data sheets, requirements documents, assembly instructions," marketing materials, and other guidance that instructs and encourages customers and end-users to operate the VUSION platform in a manner that infringes the patents (Compl. ¶¶ 25-26, 33-34).
  • Willful Infringement: The complaint alleges that Defendants have knowledge of the asserted patents "at least as of the date when they were notified of the filing of this action." This forms the basis for a claim of post-suit willful infringement (Compl. ¶¶ 28, 36).

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

  1. An Architectural Question: For the '321 patent, a central issue will be whether the accused VUSION platform possesses the specific, two-part distributed architecture required by the claims. The case may turn on whether the system functionally separates "object detection" from "video analytics" and sends only "extracted portions of frames" to a remote system that is "operated by a different business entity" than the on-site cameras.
  2. A Methodological Question: For the '272 patent, a key evidentiary question will be one of process equivalence: does the accused system build its product knowledge base by performing the specific claimed method of "cross-correlating... bounding boxes across one or more stores" to "build training data," or does it use a different, non-infringing methodology for product recognition?
  3. An Evidentiary Question: As the complaint provides only high-level, conclusory allegations of infringement and does not include the referenced claim charts, a threshold issue for the litigation will be what evidence emerges during discovery to substantiate the claim that the accused VUSION platform actually practices each of the specific technical and structural limitations of the asserted patents.