Digital Commons Network^™

Automatic Insertion Of Hot Keywords To Drive Traffic On Advertisements, Yunruo Liang, Xi Li, Weiwei Ji, Meng Meng, Zhouxiang Huang, Yiting Cheng, Chien-Ju Chang, Xue Yang, Rongxi Cai

Defensive Publications Series

Product titles and descriptions that include appropriate keywords, when used in an online advertisement, can improve the shopping feed quality and resultant traffic to the advertiser. However, online merchants lack knowledge of currently trending or popular keywords, and lacking keyword ideation, may choose suboptimal product titles. This disclosure describes techniques that enable online merchants to automatically optimize product titles or descriptions, e.g., as used in online ads. Trending or popular keywords relevant to the product are automatically added to the product title or description. Unique, product-specific insights gleaned from searched terms are utilized to improve title effectiveness automatically and at …

Using Eye Gaze For Interactive Computing, Alexander James Faaborg, Michael Schoenberg

Defensive Publications Series

User input to desktop and laptop computers is largely via the keyboard, pointing devices such as a mouse/trackpad, and to some extent, the touchscreen. Thus far, using eye gaze as an input, based on data from on-device cameras (e.g., laptop or desktop cameras) requires calibration steps; even so, since such cameras are relatively far from the user, the resulting input lacks precision. This disclosure describes techniques for human-computer interaction based on eye gaze derived from the user’s smart glasses. With user permission, the user’s eye movements, as captured by the camera on the user’s smart glasses, provides an additional interactive …

Electrical Conductive Undercarriage Protection And Wiper Blade, Hp Inc

Defensive Publications Series

This idea describes how making conductive parts that touch and surround printhead nozzle plate can help on improving image quality of prints.

Intuitive Modeling Of Voxel-Based 3d Objects, Luis Quesada Torres

Defensive Publications Series

Voxels (volumetric pixels) are virtual cubes that can be used to construct, model, or analyze virtual three-dimensional objects. Voxels find extensive use in video game design, animation, three-dimensional printing, etc. Software modeling of voxel objects is difficult for a designer working on a two-dimensional computer screen because voxel placement in the third dimension (perpendicular to the computer screen) can be ambiguous or obscured.

This disclosure describes a simple, low-cost, and expandable hardware interface and accompanying software to intuitively model or define voxel-based 3D objects. The base has a matrix of connectors. A number of cubic pieces or physical voxels of …

Air Pressure System Assisted Prime From Ink Pump, Hp Inc

Defensive Publications Series

The purge process of the Ink Delivery System (IDS) of a printer is a critical process

from several points of view.

It is one of the process of the Out of the Box Experience, it is the first contact of the

customer with the printer that has just been bought. And the first time the printer starts

moving ink. So, any problem related to the purge has a huge impact on customer

experience.

Intent-Based Networking From The Iot Edge To The Application Server, Tim Szigeti, Robert Barton, Jerome Henry, Dave Zacks

Defensive Publications Series

The dynamic management of traffic within an operational technology (OT) network raises a number of challenges. To address those types of challenges, techniques are presented herein that enable end-to-end intent-based networking to control access between the OT domain and on-premise or cloud-based data center (DC) domains. Aspects of the presented techniques employ deep packet inspection (DPI) of industrial protocols within the OT domain (e.g. by sensors) and map Internet of Things (IoT) devices and traffic flows to abstract tags (through, e.g. a robust security facility), export such tags to a common policy server that bridges both domains, assign the IoT …

Determining Product Complementarity By Analyzing Search Query Streams, Luis Benavides, Timothy Shur, Anshul Sawant, Sreenivas Gollapudi, Nishith Parikh, Akshay Gupta, Leon Xu

Defensive Publications Series

Determining complementarity between products when viewing products online, e.g., via a shopping website or a search engine, is not straightforward. The information required to determine product complementarity is unstructured and/or spread across multiple sources. While co-purchase data can be used to infer complementarity, this requires large volumes of purchase data that is often unavailable. This disclosure describes techniques to identify relationships between products based on analysis of web search queries, subsequent clicks or other user actions, obtained with user permission. The techniques rely on the observation that query streams often include searches for related products, e.g., in the form of …

Two-Handed Typing Method On An Arbitrary Surface, Shumin Zhai, Mingrui Zhang

Defensive Publications Series

A computing device may detect user input, such as finger movements resembling typing on an invisible virtual keyboard in the air or on any surface, to enable typing. The computing device may use sensors (e.g., accelerometers, cameras, piezoelectric sensors, etc.) to detect the user’s finger movements, such as the user’s fingers moving through the air and/or contacting a surface. The computing device may then decode (or, in other words, convert, interpret, analyze, etc.) the detected finger movements to identify corresponding inputs representative of characters (e.g., alphanumeric characters, national characters, special characters, etc.). To reduce input errors, the computing device may …

Real-Time Secure Sharing Of Live Meeting Content Through Voice Commands Or User Clicks, Rajarshee Dhar, Deepesh Arora, Ram Mohan R

Defensive Publications Series

With the dramatic increase in working from home due to the COVID-19 situation, the reliance on online collaborative sessions has increased manifold and there is a significant need to have an enhanced immersive sharing and collaborative experience in live meetings. Existing sharing and collaboration mechanisms (including the sharing of meeting presentations ahead of time, etc.) all suffer from different weaknesses. To address these types of challenges, techniques are presented herein that support a means whereby, among other things, an authorized participant in an online meeting may request (through, for example, clicking on a shared meeting screen, using a voice input, …

Finding Match Avoidance Attempts At Scale With Video Expansion, Roman Vorushin, Philipp Neubeck, Hanna Maria Pasula, Filip Pavetić, Conrad Burchert

Defensive Publications Series

An important objective of user-generated content platforms such as audio/video hosting or streaming platforms is to ensure that content that is available via their platforms is authorized for use, e.g., is provided by the true owner or with due permission of the true owner. To ensure that unauthorized content is not made available, such platforms match uploaded videos against a repository of reference (original) videos. To avoid video content being matched, content uploaders utilize constantly evolving new content transformation strategies when uploading unauthorized content. This disclosure describes automated techniques that help speed up and scale the collection of training examples …

Withholding Content And Recommendation Delivery To Avoid Spoilers, Alexander James Faaborg, Brett Barros

Defensive Publications Series

Online news portals, personalized feeds, etc. can include content spoilers that can ruin a user’s online experience by revealing details of the plot of content such as television shows, movies, etc. This disclosure describes techniques to avoid spoilers by delaying the presentation of material pertaining to various types of content until a user has consumed the content. With user permission, a user’s content consumption history is accessed and utilized to identify news articles and other media that can be a content spoiler. Analysis of articles and other media can be performed using machine learning models or other suitable techniques to …

Vacuum Nozzle For Metal Powder Retrieval, Hp Inc

Defensive Publications Series

This disclosure relates to the process of recovering the non-solidified powder in additive manufacturing

systems using binder jetting and powder bed fusion technologies. Typically, the powder recovery from

the build platforms is a manual operation that consists of aspirating the powder particles through a

nozzle. The powder is transported to a tank using a vacuum system.

With the current set of nozzles, aspirating the powder is slow and labor-intensive. The material

transport pipes are often clogged, and the nozzles are stuck against the build platforms surfaces. The

processing time is highly variable and dependent on the operator's ability, which is …

Test Coverage Analysis By Diffing Production Logs Against Integration Test Logs, Yifan Gao, Ben Rafferty, Tylor Sampson, Joseph Kubik

Defensive Publications Series

This disclosure describes tools and techniques to analyze input and configuration coverage of software by performing large scale analysis of production and integration test data in a privacy-compliant manner. Test coverage analysis is made possible for highly complex input schema or for input schema that evolve with time. The techniques uncover the input-subspace that has been thus far integration-tested and the input-subspace that remains untested (known as test-gaps). When gathering data on how a specific software input/configuration is used in production, the techniques determine not only if they are used, but also how often they are used. Such data can …

Slide-Based Navigation And Understanding For Video Content, Matthew Sharifi

Defensive Publications Series

In educational videos, the speaker often presents a set of slides that serve as logical cues and content markers. Currently, video platforms do not use the strong cues already available in the form of slides. This disclosure describes techniques that enable video viewers to more naturally navigate the video. With user permission, computer vision techniques are applied to video content to detect whether it includes a presentation, to track individual slide changes, and to recognize content displayed on each slide. Automatic understanding of slide content is utilized to improve speech recognition and content captioning, to enhance video search, and to …

Detection And Prevention Of Eapol-Start Flood Attacks In Lln Environments, Lele Zhang, Yajun Xia, Akram Sheriff, Chuanwei Li

Defensive Publications Series

Due to limited resources in wireless sensors, such sensors are easily attacked by flood joining messages. To address these types of challenges, techniques are presented herein that support making simple nodes (such as wireless sensors) "smarter" to allow them to identify malicious node(s) from legal neighbors. Aspects of the techniques presented herein leverage, among other things, the Extensible Authentication Protocol (EAP) over local area network (LAN), or EAPOL, protocol and employ, among other things, an EAPOL neighbor status profile, or ENSP. A node may generate an ENSP for each connection rather than for each node based on a unique address. …

Intelligent Mechanism For Robocall Prevention Using Artificial Intelligence-Driven Interactive Question And Answer Techniques, Anupam Mukherjee, Faisal Siyavudeen, Sanjay Sinha

Defensive Publications Series

A robocall is a phone call that uses a computerized auto-dialer to deliver a pre-recorded message, as if from a robot. Robocalls are often associated with political and telemarketing phone campaigns. Some robocalls use personalized audio messages to simulate an actual personal phone call. Criminals and dishonest robocallers often alter or spoof the calling number of their outbound telephone calls in order to deceive a called party. This deception increases the chance that the called party will answer a robocall. In other cases, the spam calls may be more malicious. This proposal provides an artificial intelligence (AI)-based interactive and intelligent …

Ink Consumption Forecasting During Pandemic Using Deep Learning Approach, Hp Inc

Defensive Publications Series

A novel methodology is designed to forecast ink usage for following 2 months from multivariable time

series data. Proposed approach also incorporates the impact of unprecedented events such as

COVID pandemic. It uses Deep Learning Encoder Decoder model for identifying the seasonality,

trend and sudden changepoints in multivariate time series.

Techniques To Improve The Hit Rate Of Unicast Node-To-Node (N2n) Delivery In Channel-Hopping And Multi-Hop Low-Power And Lossy Networks (Llns), Lele Zhang, Chuanwei Li, Huimin She, Yajun Xia

Defensive Publications Series

In a large scale wireless mesh network, such as a Wireless Smart Utility Network (Wi-SUN), there can be many losses. For example, a transmitter typically does not know the availability of a receiver when transmitting, which can create losses. In another example, link quality measurements, such as expected transmission count (ETX), may not be well represented for every channel. In one instance, this proposal provides an improved scheduling technique that leverages Broadcast Interval (BI) information to determine receive (RX) and transmit (TX) plans for child nodes. Child nodes can select an appropriate parent node according to received schedules, which can …

Optimizing Notifications In A Collaboration Application, Brian Powell, Robert Barton, Jerome Henry, Cesar Obediente

Defensive Publications Series

Collaboration tools typically utilize many manual settings or 'knobs', but often lack an ability to learn what features, settings, etc. may be important to a user, in general. Optimization of individual collaboration tools has been studied; however, such studies are typically focused on making an individual tool more efficient. In contrast, this proposal provides techniques for optimizing collaboration tool notifications based on user habits, real-time context, and behavior in a manner that is not 'just' based on the way in which a user utilizes a particular collaboration tool.

On-Device Query And Metadata Caching For Expedited Inference And Rendering Of Answer Cards, Keun Soo Yim

Defensive Publications Series

This publication describes methods and techniques for mobile devices to enhance the user experience associated with search applications. In aspects, an Intent Manager associates a query suggestion to a corresponding answer card (e.g., search result) for a given query text. The association may involve the Intent Manager identifying that the submission of a query suggestion results in a search application presenting one or more corresponding answer cards. Further, the association may include the production of metadata relating the query text to the submitted query suggestion. The Intent Manager can then cache the metadata on the device. In doing so, when …

Dynamically Adapting Press User Interface Based On Press Operator's Estimated Location, Hp Inc

Defensive Publications Series

This disclosure relates to the field of User Experience Design and User Interface

Design within the context of Industrial Presses. The idea is concerned with the

displays mounted in various locations on the presses, and the software

provided by press manufacturers allowing press operators to control and

operate the press.

Unwarping Pages Using Semantic Segmentation Information And Dense Image Warping Matrix, Hp Inc

Defensive Publications Series

Scanning pages from books or magazines usually result in a digital image with warped

regions, which impacts the text quality and decreases OCR precision. To address this problem,

this work proposes a method using machine learning to estimate an unwarping matrix that can

fix the warping regions of the scanned page. Furthermore, this method allows the user

interaction with the final document, making it possible to adjust the level of unwarping

adjustment.

Pcl3gui Formats Comparator To Verify Print Data, Hp Inc

Defensive Publications Series

Cloud platform enables anyone with an Internet-enabled device to print documents, photos

by rendering any document entering the system into a printable format and then sends to

the corresponding printer for printing. Document gets converted into a printable format in

the rendering engine component, the most common format being PCL3GUI (Printer

Command Language) [2]. Cloud platform supports several features. It is the rendering

engine component of the cloud where the document gets converted into a printable format,

the most common format being PCL3GUI (Printer Command Language). Cloud print

platform supports several features. Rendering features are hard to validate as it …

Machine Learning Based Distortion Correction, Kathan Shukla

Defensive Publications Series

A vehicle system may use a computer vision module to rectify distortions in images captured by one or more fish-eye cameras (hereinafter referred to as “fish-eye camera images”) of the surroundings of a vehicle (e.g., an automobile, a motorcycle, a bus, a recreational vehicle (RV), a semi-trailer truck, a tractor or other type of farm equipment, train, a plane, a boat, a helicopter, a personal transport vehicle, etc.). In some examples, a machine-learning model (hereinafter referred to as “ML model”) may calibrate (or, in other words, tune) intrinsic parameters (e.g., the fish-eye camera images’ focal length ((x,y) coordinates), principal point …

Carrie 2.0: Insight Extraction From Unstructured Data Using Proximity Clustering Engine, Hp Inc

Defensive Publications Series

CaRRIE 2.0 extracts key insights/summaries from unstructured data such as papers, news

document, notes, blog scrapped website, and RFP (Request for Proposal) data. The user gains

faster understanding and insights to make informed decisions. The engine groups/ranks

collections of documents based on keywords of interest presented within the documents. The

user can then focus on a handful of documents aligned with their interests, reducing the time

needed to read each document and create insights. He/she can also understand the larger data

trends without needing to read the entire stack of documents, one by one. We accomplish those

tasks by stacking …

Method And System For Managing Data Flow Between Source System And Target System, Shreyas Kunjal Chandrahas

Defensive Publications Series

The present disclosure relates to a method and a system (103) for managing data flow between a source system (101) and a target system (105). Here, the system comprises a Metadata Definer (MD) 201 for defining metadata of the input dataset associated with the source system (101), output dataset associated with the target system (105), the data transfer pipeline between the input and output datasets and its internal transformation. The system further comprises a Data Flow Manager (DFM) (203) which extracts metadata input from the MD 201, and determines source type, target type, transfer type, data quality strategy, frequency of …

Peer Learning On The Edge In Vehicles, Hanumant Prasad R. Singh

Defensive Publications Series

A vehicle head unit may train a surround-view (SV) detection module to rectify distortions in fish-eye camera images of the surroundings of a vehicle by comparing the object (e.g., traffic signs, lane markings, etc.) detection results of the SV detection module with those of an advanced driver assistance system (ADAS) detection module (e.g., while the SV detection module and the ADAS detection module are detecting the same objects of the same scenery). The vehicle head unit may receive the object detection results of the ADAS detection module by using one or more communication processes. For example, the vehicle head unit …

Evaluation And Verification Of Video Editing Output Based On Content Fingerprinting, Christian Ingemar Falk, Ekaterina Starostina, Karol Rozycki

Defensive Publications Series

Modifications to video processing pipelines, e.g., editlist generators, can cause the output videos to not match a user specification. This disclosure describes techniques for the automated verification of revisions to video editlist generators and/or video edit engines. A random set of edited videos and the corresponding edit specifications are selected (205). The number of videos in the set is determined based on an estimation of the likelihood of an error in the output video as a result of changes in the editlist generator. Editlists are generated for each selected video based on a control editlist generator and a modified editlist …

Mobile Head-Up Display, Sam Lin, Dexter (Chia Hang) Yeh

Defensive Publications Series

Wearable displays such as smartglasses are heavy and performance-constrained due to weight restrictions. This disclosure describes a mobile head-up display (MHUD) comprising a lightweight component that sits on the user’s face that is connected to a smartphone or other device carried on the user’s person. Compute-intensive, heavy components are offloaded to and leverage the existing capabilities of the smartphone. Components that deliver a multimedia user experience, e.g., near-eye display, camera, sensors, headphones, etc. are located on the head-up display. The app-developer ecosystem of the smartphone is leveraged to provide MHUD apps at scale and competitive cost.

Smooth Transitioning Between Augmented And Virtual Reality, Chris Ross

Defensive Publications Series

This disclosure describes techniques that enable a user to start in an augmented reality (AR) environment (where virtual objects can be observed in their physical surroundings) and move to a virtual reality (VR) environment (where virtual objects alone can be examined from various perspectives in detail or from a distance). The user can make the reverse VR-to-AR journey as well. The AR↔VR transitioning and the examination of objects from various perspectives is made possible by smooth transformations of scale and of rotation between the coordinates of the world (e.g., the scene) and the object.