2:26-cv-00847
NEC Corp v. Anker Innovations Technology Co Ltd
I. Executive Summary and Procedural Information
- Parties & Counsel:
- Plaintiff: NEC Corporation (Japan)
- Defendant: Anker Innovations Technology Co., Ltd. (People's Republic of China) and Anker Innovations Ltd. (Hong Kong)
- Plaintiff’s Counsel: Patton Tidwell & Culbertson, LLP; Mayer Brown LLP
- Case Identification: 2:26-cv-00847, E.D. Tex., 09/18/2026
- Venue Allegations: Venue is alleged to be proper in the district because the defendants are not residents of the United States and may therefore be sued in any judicial district pursuant to 28 U.S.C. § 1391(c)(3).
- Core Dispute: Plaintiff alleges that Defendant’s "eufy" brand of smart surveillance products infringes six U.S. patents related to video analysis, object recognition, and surveillance system control.
- Technical Context: The technology at issue resides in the domain of AI-powered video surveillance, which includes intelligent object detection, facial recognition, and user interface controls for smart security systems.
- Key Procedural History: The complaint indicates a prior history between the parties, including a letter sent by NEC to Anker on June 3, 2024, offering to license four of the six patents-in-suit. The complaint also references a prior patent infringement complaint filed by NEC against Anker on September 3, 2024, suggesting this filing is part of an escalating dispute. Additionally, the complaint notes a 2025 settlement between Anker and the State of New York concerning security vulnerabilities in its eufy home security products.
Case Timeline
| Date | Event |
|---|---|
| 2012-07-31 | '315 Patent Priority Date |
| 2013-05-29 | NEC-led consortium awarded Singapore Safe City Test Bed Project |
| 2014-04-28 | '713 Patent Priority Date |
| 2014-09-16 | '619 Patent Priority Date |
| 2015-03-27 | '776 Patent Priority Date |
| 2017-03-17 | '336 Patent Priority Date |
| 2017-03-30 | '985 Patent Priority Date |
| 2019-03-05 | '619 Patent Issue Date |
| 2020-02-04 | '713 Patent Issue Date |
| 2020-07-07 | '336 Patent Issue Date |
| 2021-10-12 | '776 Patent Issue Date |
| 2024-06-03 | NEC sends pre-suit notice letter to Anker |
| 2024-09-03 | NEC files prior patent infringement complaint against Anker |
| 2025-09-23 | '985 Patent Issue Date |
| 2026-05-26 | '315 Patent Issue Date |
| 2026-09-18 | Complaint Filing Date |
II. Technology and Patent(s)-in-Suit Analysis
U.S. Patent No. 12,423,985 - “Information Processing Apparatus, Control Method, and Program”
- Issued: September 23, 2025
The Invention Explained
- Problem Addressed: The patent's background describes shortcomings in prior art security systems with respect to detecting objects, such as unattended luggage, that could pose a security risk Compl. ¶35 ’985 Patent, col. 1:50-55
- The Patented Solution: The invention is an information processing system that performs a two-stage analysis. First, it detects a "target object" (e.g., a package) in a video segment and outputs a corresponding image. Subsequently, it detects a "person" within a "vicinity region" of that same object in a later video segment and outputs a second image showing both the person and the object Compl. ¶34 ’985 Patent, abstract This allows the system to correlate the appearance of an object with subsequent human interaction Compl. ¶36 ’985 Patent, col. 4:1-20
- Technical Importance: This technology enables more sophisticated event analysis than simple motion detection, allowing a system to differentiate between events like a package delivery and a potential package theft by linking an object to a person over time Compl. ¶¶34-36
Key Claims at a Glance
The complaint asserts infringement of claims 1-30 Compl. ¶36 Independent claim 1 of the '985 Patent recites the following essential elements:
- detecting a target object in a first video segment;
- selecting a first video frame from the first video segment, wherein the first video frame includes at least a part of the target object;
- outputting a first image based on the first video frame, wherein the first image includes at least a part of the target object;
- detecting a person in a second video segment, wherein the person is within a vicinity region;
- selecting a second video frame from the second video segment, wherein the second video frame includes at least a part of the target object and at least a part of the person;
- outputting a second image based on the second video frame, wherein the second image includes at least a part of the target object and at least a part of the person;
- wherein the second video segment occurs chronologically after the first video segment.
U.S. Patent No. 10,223,619 - “Video Monitoring Apparatus, Control Apparatus, Control Method, and Non-transitory Readable Storage Medium”
- Issued: March 5, 2019
The Invention Explained
- Problem Addressed: The patent notes that conventional video analysis systems were limited to detecting simple behaviors, such as a person entering a prohibited area, and were unable to detect more "complicated behavior" Compl. ¶53 ’619 Patent, col. 1:64-67
- The Patented Solution: The invention describes a system comprising storage for a first and second condition, input circuits for receiving those conditions, and a conditional operator input circuit for defining a "logical relationship" between them. An output circuit then signals when the defined logical relationship between events satisfying the conditions is met Compl. ¶52 ’619 Patent, abstract This structure allows a user to create complex, multi-part rules for surveillance alerts ’619 Patent, col. 3:42-47
- Technical Importance: The technology provides users with customizable tools to define complex surveillance logic, moving beyond simple triggers to create conditional alerts based on multiple events, thereby potentially reducing false positives and capturing more specific events of interest Compl. ¶55
Key Claims at a Glance
The complaint asserts infringement of claims 1, 7-11, 15, and 16 Compl. ¶56 Independent claim 1 of the '619 Patent recites the following essential elements:
- a first storage which stores a first condition for detecting a first event from a video;
- a second storage which stores a second condition for detecting a second event from the video;
- an output circuit which outputs a signal indicating that a logical relationship designated in advance by the first event and the second event is satisfied;
- a first input circuit which receives an input of the first condition;
- a second input circuit which receives an input of the second condition; and
- a conditional operator input circuit which receives an input of a conditional operator indicating the logical relationship.
U.S. Patent No. 12,641,315 - “Image Processing System, Image Processing Method, and Program”
- Issued: May 26, 2026
Technology Synopsis
The patent addresses problems associated with tracking people across multiple cameras, noting that human involvement complicates the monitoring task Compl. ¶135 The invention provides a system that receives video from different cameras and, when a second person is detected on a different camera while a first person is being displayed, outputs a notification and allows a user to select different camera views Compl. ¶134 Compl. ¶137
Asserted Claims
Claims 1-18 are asserted Compl. ¶138
Accused Features
The complaint alleges that the Accused Products' multi-camera environment, which receives and processes video from different cameras to detect and display multiple people and allows users to select between camera views, infringes the '315 Patent Compl. ¶139
U.S. Patent No. 10,706,336 - “Recognition In Unlabeled Videos With Domain Adversarial Learning and Knowledge Distillation”
- Issued: July 7, 2020
Technology Synopsis
The patent is directed to solving problems with object recognition in unlabeled videos by improving the accuracy of training models Compl. ¶92 The invention describes a system that pre-trains a recognition engine using convolutional neural networks (CNNs) on a domain of labeled still images and then adapts that engine to a video domain by applying it to synthetically degraded images and random unlabeled video frames Compl. ¶91 Compl. ¶94
Asserted Claims
Claims 1, 9, 11, 12, 15, 19, and 20 are asserted Compl. ¶95
Accused Features
The accused features are Anker's "AI Detection" and "Self-learning" capabilities, which allegedly employ a "LOCAL AI deep learning neural network" to recognize objects and continuously improve accuracy by learning from target recognitions and using self-learning algorithms Compl. ¶¶95, 99, 100
U.S. Patent No. 10,552,713 - “Image Analysis System, Image Analysis Method, and Storage Medium”
- Issued: February 4, 2020
Technology Synopsis
The patent aims to overcome deficiencies in conventional systems for tracking suspicious persons, particularly when the monitored area is wide or the subject is not registered in advance Compl. ¶116 The invention generates an "appearance record" of an object from captured images—including time and location information—and provides an output identifying the object when its appearance pattern matches a predetermined pattern Compl. ¶113 Compl. ¶118
Asserted Claims
Claims 1-6 and 11 are asserted Compl. ¶119
Accused Features
The accused feature is the "Cross-Camera Tracking" function, which allegedly uses a "human-feature-recognition algorithm" to identify individuals across multiple cameras, merge the footage into a single video story, and provide an output identifying the monitored person Compl. ¶120
U.S. Patent No. 11,144,776 - “Mobile Surveillance Apparatus, Program, and Control Method”
- Issued: October 12, 2021
Technology Synopsis
The patent addresses the problem of "poor operability" in mobile surveillance applications where the same touch input (e.g., a slide operation) could be used for different processes, creating ambiguity for the user Compl. ¶73 The invention provides a mobile apparatus that can switch between a first process (e.g., setting an event detection region) and a second process (e.g., changing the display range) in response to a specific touch operation performed in a predetermined order Compl. ¶72
Asserted Claims
Claims 1-16 are asserted Compl. ¶76
Accused Features
The accused functionality is the "activity zone detection" feature within the eufy mobile application, where users can perform touch operations to either set a detection region or change the display range (pan/zoom) of the surveillance image Compl. ¶¶76, 79, 82
III. The Accused Instrumentality
Product Identification
The accused instrumentalities are a wide array of Anker's "eufy" branded smart home security products, including the eufy Video Doorbell, eufyCam, SoloCam, and HomeBase models, as well as associated software such as the "eufy Security Application" and the "BionicMind AI Service" Compl. ¶29 Compl. ¶36 Compl. ¶56 Compl. ¶76 Compl. ¶95 Compl. ¶119 Compl. ¶138
Functionality and Market Context
The accused products are internet-connected security cameras, doorbells, and local storage hubs that capture video and use AI-driven software to perform intelligent analysis (Compl. ¶¶34, 99). Key functionalities alleged in the complaint include "Delivery Guard," which specifically monitors for package deliveries and subsequent human interaction Compl. ¶39; "AI Detection" for identifying humans, faces, pets, and vehicles Compl. ¶20; "Cross-Camera Tracking" for monitoring individuals across multiple cameras Compl. ¶120; and various customizable rules and "Activity Zones" set by the user through a mobile application Compl. ¶60 One screenshot from the eufy app shows an event log listing "package detection with a part of a person," illustrating the system's ability to correlate object and person detection Compl. p. 21 The complaint positions Anker as a major market participant that has achieved significant growth, allegedly aided by anti-competitive practices Compl. ¶31
IV. Analysis of Infringement Allegations
'985 Patent Infringement Allegations
| Claim Element (from Independent Claim 1) | Alleged Infringing Functionality | Complaint Citation | Patent Citation |
|---|---|---|---|
| detecting a target object in a first video segment | The accused products' "Delivery Guard," "Package Detection," or "Pet Detection" features use AI to detect objects like packages or pets in a video stream. | ¶39 | col. 4:1-5 |
| outputting a first image based on the first video frame, wherein the first image includes at least a part of the target object | The system generates a notification and a video clip or image of the initial event, such as a package being delivered. A screenshot of the eufy app shows a "Package Detection" event in the log Compl. p. 19 | ¶39 | col. 4:9-12 |
| detecting a person in a second video segment, wherein the person is within a vicinity region | The "Delivery Guard" feature is described as notifying the user "when someone approaches your package." | ¶39 | col. 4:13-16 |
| outputting a second image based on the second video frame, wherein the second image includes at least a part of the target object and at least a part of the person | The system generates a subsequent notification and image showing the person who has approached the previously detected object. The complaint provides an app screenshot showing an event described as "package detection with a part of a person" Compl. p. 21 | ¶39 | col. 4:17-20 |
| wherein the second video segment occurs chronologically after the first video segment | The system first detects the object and later detects the person approaching it. | ¶34 | col. 4:21-23 |
- Identified Points of Contention:
- Scope Questions: A question may arise as to whether the term "target object", which the patent describes in the context of "unattended baggage" '985 Patent, col. 1:50-55, can be construed to cover the broad range of objects the accused products allegedly detect, including animate pets and vehicles.
- Technical Questions: The infringement analysis may focus on the definition of "vicinity region". The complaint does not specify how this region is defined by the accused products, and a dispute could arise over whether the accused system's method of detecting a person "approaching" a package meets the specific technical requirements for detecting a person "within a vicinity region" as potentially narrowed by the patent's specification.
'619 Patent Infringement Allegations
| Claim Element (from Independent Claim 1) | Alleged Infringing Functionality | Complaint Citation | Patent Citation |
|---|---|---|---|
| a first storage which stores a first condition for detecting a first event from a video | The accused products allow users to define a first condition, such as setting an "Activity Zone" to detect motion only in a specific area. | ¶60 | col. 10:8-12 |
| a second storage which stores a second condition for detecting a second event from the video | The accused products allow users to define a second, distinct condition, such as a "Detection Type" (e.g., "Human Only") or a "Security Mode" (e.g., "Away Mode"). | ¶60 | col. 10:13-16 |
| an output circuit which outputs a signal indicating that a logical relationship...is satisfied | The system sends a notification or triggers an alarm only when the combination of user-defined conditions is met (e.g., a human is detected inside the activity zone while the system is in away mode). | ¶58 | col. 10:17-21 |
| a first input circuit which receives an input of the first condition; a second input circuit which receives an input of the second condition; and a conditional operator input circuit which receives an input of a conditional operator indicating the logical relationship | The eufy Security App provides a user interface for setting up "Activity Zones," "Security Modes," and "Detection Sensitivity," which collectively define the conditions and the logical relationship for triggering an alert. A screenshot shows the "Delivery Guard" settings page where multiple conditions can be enabled Compl. p. 37 | ¶59 | col. 10:59-67 |
- Identified Points of Contention:
- Technical Questions: A central question will be whether the accused products' user interface for combining various settings (e.g., detection type and activity zone) constitutes the "conditional operator input circuit" that receives a "conditional operator indicating the logical relationship" as claimed. The defense may argue that combining filters is functionally different from the patent's disclosure of explicit logical operators like "AND, OR, ... THEN" '619 Patent, col. 3:42-47
- Scope Questions: The parties may dispute whether a single detected motion that is checked against multiple criteria (e.g., is it a 'person' AND is it in 'Zone A'?) satisfies the claim language of a "first event" and a "second event," or if the patent requires two separate and distinct events to occur.
V. Key Claim Terms for Construction
For the '985 Patent:
- The Term: "vicinity region"
- Context and Importance: This term is critical for determining infringement, as it defines the spatial relationship required between the detected "target object" and the subsequently detected "person." A narrow definition could allow Defendant to argue non-infringement if its system does not define a specific, bounded area around the object. Practitioners may focus on this term because its construction will dictate the type of spatial evidence NEC must produce.
- Intrinsic Evidence for Interpretation:
- Evidence for a Broader Interpretation: The patent abstract describes detecting a person "in a vicinity of the target object," suggesting a general, context-dependent area rather than a rigidly defined one '985 Patent, abstract
- Evidence for a Narrower Interpretation: The specification discloses an embodiment where the vicinity region is set "so as to surround the target object" and is illustrated as a defined geometric shape in a figure '985 Patent, col. 6:18-20 ’985 Patent, Fig. 6A This could support a narrower construction requiring a specifically configured zone.
For the '619 Patent:
- The Term: "logical relationship"
- Context and Importance: This term is the core of the invention, which purports to enable the detection of "complicated behavior" by combining simple events. The case may turn on whether Anker's system of applying multiple user-selected filters (e.g., "Human Only" and an "Activity Zone") implements a "logical relationship" as that term is used in the patent.
- Intrinsic Evidence for Interpretation:
- Evidence for a Broader Interpretation: The claim requires only that the relationship be "designated in advance," which could be argued to encompass any pre-set combination of rules that results in a conditional alert '619 Patent, claim 1
- Evidence for a Narrower Interpretation: The specification provides an explicit, non-limiting list of formal logical operators: "AND, OR, NOR, THEN, NAND, XOR, and the like" '619 Patent, col. 3:42-47 This provides strong evidence for a narrower construction that may require the accused system to implement one of these specific logical functions, rather than just a layered filtering process.
VI. Other Allegations
- Indirect Infringement: For all asserted patents, the complaint alleges induced infringement, stating that Defendant’s user manuals, websites, and tutorial videos instruct customers on how to use the accused features, such as setting up Delivery Guard, Activity Zones, and Cross-Camera Tracking Compl. ¶¶41-42 Compl. ¶62 Compl. ¶82 Compl. ¶103 Compl. ¶124 Compl. ¶143 The complaint also alleges contributory infringement on the basis that components of the accused products are especially made for practicing the patents and are not staple articles of commerce (Compl. ¶¶46-47; Compl. ¶66-67; Compl. ¶85-86; Compl. ¶107-108; Compl. ¶128-129; Compl. ¶147-148).
- Willful Infringement: The complaint alleges willful infringement for all six patents. For the '619, '336, '713, and '776 patents, it alleges Anker had pre-suit knowledge based on a notice letter dated June 3, 2024 Compl. ¶68 Compl. ¶87 Compl. ¶109 Compl. ¶130 For all patents, it alleges knowledge as of the filing of this or a prior complaint, and that Anker's continued infringement is despite an objectively high likelihood that its actions constitute infringement of valid patents Compl. ¶49 Compl. ¶69 Compl. ¶88 Compl. ¶110 Compl. ¶131 Compl. ¶150
VII. Analyst’s Conclusion: Key Questions for the Case
- A central issue for multiple patents ('619, '776, '985) will be one of functional implementation: Does the accused eufy system's method of combining user-configurable filters (e.g., "Human Only" + "Activity Zone") perform the specific, structured processes required by the claims, such as applying a formal "logical relationship" ('619 patent) or switching between discrete operational modes based on a predetermined touch order ('776 patent)?
- A key evidentiary question for the AI-focused patents ('336, '713) will be one of algorithmic equivalence: Does Anker's proprietary "BionicMind AI" and "Cross-Camera Tracking" technology operate in a manner that maps onto the specific patented methods for adapting a recognition engine using synthetically degraded images ('336 patent) and generating time-and-location-based "appearance records" to track subjects ('713 patent)?
- A core issue will be one of definitional scope: Can claim terms rooted in specific examples in the patent specifications, such as "target object" described in the context of "unattended baggage" ('985 patent), be construed broadly enough to cover the diverse range of functions offered by the accused products, which include detecting people, pets, and vehicles?