Abstract: The present application discloses a video-based human behavior recognition method, apparatus, device and storage medium, and relates to the technical field of human recognitions. The specific implementation scheme lies in: acquiring a human rectangle of each video frame of the video to be recognized, where each human rectangle includes a plurality of human key points, and each of the human key points has a key point feature; constructing a feature matrix according to the human rectangle of the each video frame; convolving the feature matrix with respect to a video frame quantity dimension to obtain a first convolution result and convolving the feature matrix with respect to a key point quantity dimension to obtain a second convolution result; inputting the first convolution result and the second convolution result into a preset classification model to obtain a human behavior category of the video to be recognized.
Abstract: A printing apparatus connectable to an external power supply having outputtable voltage information, including: a printing head performing printing by electric power supplied from the external power supply; a resistor having an electric resistivity based on the printing head; a detector detecting a substantial voltage; and a controller configured to: obtain the voltage information from the external power supply; decide a tentative voltage; request the external power supply to output the tentative voltage; energize the resistor with the tentative voltage; decide the substantial voltage if a voltage corresponding to each of the obtained voltage information is supplied based on a result of the detector during the energizing of the resistor; decide an electric power supply voltage based on the substantial voltage; and request the external power supply to output the electric power supply voltage.
Abstract: A cabinet includes a first area and a second area arranged in a horizontal direction, an image forming apparatus, and a reading device. The image forming apparatus is accommodated in the first area and forms an image on a sheet and ejects the sheet with the image. The reading device is accommodated in the second area of the cabinet and reads an image of a document. At least a part of the reading device is pulled out from an internal space of the cabinet in a direction orthogonal to the horizontal direction in which the first area and the second area are arranged in a horizontal plane.
Abstract: Certain aspects relate to systems and usage techniques for modular lateral flow assay reader devices that can receive a number of different modules having a barcode scanning input device and optional network connectivity capabilities. A barcode scanning module can provide a simple input method that reduces errors compared to manual data entry. A network connectivity module can enable transmission of test results over a public network for standardizing, tracking and electronically connecting test results from assay reader devices located throughout a network. Such devices can programmatically implement a simplified workflow whereby pressing a single button readies the device for imaging, analyzing, and data storage/transmission and, in some implementations, configures the device to operate in one of a plurality of device operation modes.
Type:
Grant
Filed:
December 17, 2018
Date of Patent:
July 26, 2022
Assignee:
Becton, Dickinson and Company
Inventors:
Tong Xie, Stephan Hengstler, Vincent C. Moyer, Benny Wing Hung Lai
Abstract: Systems, apparatuses, and methods are directed toward determining emotional states of a subject. For example, the systems, apparatuses, and methods determine one or more previous characteristics of a pupil of an eye based on image data from eye tracking sensors, determine one or more current characteristics of the pupil based on image data from the eye tracking sensors, where the one or more current characteristics include one or more of a current position of the pupil or a current size of the pupil, track movements, based on image data from the eye tracking sensors, in a position of the pupil and changes in a size of the pupil over a time period, automatically predict an emotional state based on a comparison between the one or more previous characteristics and the one or more current characteristics, the tracked movements and the tracked changes in the size and adjust a parameter of one or more hardware devices based on the predicted emotional state.
Type:
Grant
Filed:
May 1, 2020
Date of Patent:
July 19, 2022
Assignee:
ACCENTURE GLOBAL SOLUTIONS LIMITED
Inventors:
Laetitia Kameni, Richard Vidal, Emmanuel Viale
Abstract: The present disclosure relates to an image labeling system, comprising: a support framework, the support framework being a polyhedral structure constructed by a plurality of support boards; at least two cameras mounted on at least two support boards of the plurality of support boards of the support framework; and a labeling device configured to receive two or more images to be processed taken by the at least two cameras, and to label the hand in the image to be processed.
Abstract: A series of captured images of a user is received. Using a processor, the images are processed to identify a portion of each of the images corresponding to the user. Parameters of a predetermined three-dimensional human model are modified to fit a modified version of the predetermined three-dimensional human model across the identified portions of the images to determine a set of specific parameters representing a body profile of the user.
Abstract: An electronic device for optical character recognition (OCR) and method of operation of the electronic device. The electronic device for optical character recognition comprises a memory and a processor electrically connected with the memory. The processor is configured to store, in the memory, at least one content and at least one word position table individually corresponding to the at least one content, obtain an image for at least one character scanned on a printed document by an electronic pen, identify a content corresponding to the image from the at least one content, identify a word position table corresponding to the identified content from the at least one word position table, and obtain resultant information recognized by from the optical character recognition on the image based on the identified word position table and original data of the identified content. Other various embodiments are also possible.
Type:
Grant
Filed:
October 1, 2020
Date of Patent:
July 5, 2022
Assignee:
Samsung Electronics Co., Ltd.
Inventors:
Daeyong Lee, Heejoong Kim, Dakyung Lee, Hyundong Chang, Heedae Tak
Abstract: An emotion inference device includes: an obtainer that obtains a second evaluation value group and then a first evaluation value group each including evaluation values corresponding to emotion attributes of classified emotions of a subject; a processor that, when a difference under the same emotion attribute between corresponding evaluation values in the first and second evaluation value groups does not exceed a predetermined criterion, performs a correction process to reduce the evaluation value corresponding to the same emotion attribute in the first evaluation value group, and performs an inference process to infer the emotion of the subject based on the corrected first evaluation value group, in which an emotion attribute having a relatively higher evaluation value is inferred relatively more strongly as the emotion of the subject; and an outputter that outputs a result of the inference process performed by the processor.
Abstract: A method for evaluating image quality is provided. The method may include: obtaining an image, the image including a plurality of elements, each element of the plurality of elements being a pixel or voxel, each element having a gray level; determining, based on a maximum gray level of the plurality of elements, one or more thresholds for segmenting the image; determining one or more sub-images of a region of interest by segmenting, based on the one or more thresholds, the image; and determining, based on the one or more sub-images of the region of interest, a quality index for the image.
Type:
Grant
Filed:
May 21, 2021
Date of Patent:
June 21, 2022
Assignee:
SHANGHAI UNITED IMAGING HEALTHCARE CO., LTD.
Abstract: A method for steganographic processing and compression of image data with negligible information loss, wherein said image data comprises noise and information. The method includes the steps of: acquiring image data to be processed and/or compressed, storing and/or transmitting the image data, and preparing the acquired image data for compression. The preparation step includes determining an input noise model and corresponding parameters adapted to reflect noise created by an image sensor used to capture said image data, and removing, with negligible information loss, noise from the acquired image data by use of the input noise model such as to produce noise-reduced image data.
Abstract: A gaze point estimation processing apparatus in an embodiment includes a storage configured to store a neural network as a gaze point estimation model and one or more processors. The storage stores a gaze point estimation model generated through learning based on an image for learning and information relating to a first gaze point for the image for learning. The one or more processors estimate information relating to a second gaze point with respect to an image for estimation from the image for estimation using the gaze point estimation model.
Abstract: A control portion performs a predetermined detection process on a first image from a first reading portion and a second image from a second reading portion. When the detection process on the each image is successful, the control portion performs, based on a value of an image processing variable specified from a detection result, predetermined image processing on the each image, When the detection process on the image of one side fails, a value of an image processing variable for the image of one side is specified based on a value of an image processing variable specified from the detection result for the image of the other side for which detection is successful. Based on the specified value of the image processing variable, image processing is performed on the image of one side.
Abstract: Systems and methods for gesture recognition based on depth information from a camera include, at an electronic device having a camera system, capturing a video frame and depth information associated with the video frame, identifying a foreground portion of the video frame based on the depth information, and determining whether the foreground portion matches a predefined gesture in a database. In accordance with a determination that the foreground portion matches a predefined gesture in the database, the device determines whether one or more subsequent video frames matches the one or more predefined gestures to produce a recognized gesture.
Type:
Grant
Filed:
November 13, 2018
Date of Patent:
May 24, 2022
Assignee:
Zyetric Gaming Limited
Inventors:
Pak Kit Lam, Peter Han Joo Chong, Xiang Yu
Abstract: Deterioration of communication is prevented in a communication apparatus, in which an infrastructure mode and a P2P mode are independently and concurrently executable, according to an embodiment of the present invention. Hence, in a case where the communication apparatus is in a state in which a first mode of executing wireless communication with a communication partner apparatus via an external access point and a second mode of executing wireless communication with an external apparatus without using the external access point are concurrently executable, the following selection is made. That is, a first channel width which is not accompanied by network detection processing is selected.
Abstract: An image reading apparatus includes: a reading unit that reads an image of a document placed on a platen for placing documents; and a display unit that displays part of the image of the document read by the reading unit in a range adjacent to a range, in which the document is placed, of the platen.
Abstract: Previous information generating devices or printers were designed to print a paper copy of data and/or receipts which causes a disconnect with modem day data manipulation. This information redirector can redirect generated data to and receive data from the internet and hand held devices which will open up extremely fast data exchange and data manipulation for consumers, cities, states and the federal government without the expense of having to purchase complete new systems. By simply adding the redirector we can connect all old computer systems with modem day systems that now have the ability to manipulate data automatically.
Type:
Grant
Filed:
October 28, 2019
Date of Patent:
May 3, 2022
Inventors:
Mary Lynn Sherwood, Bruce Lawrence Hall
Abstract: A bending detection device includes a processor configured to receive a read image obtained by reading a document, and notify a user of a bend in the document based on a determination that the read image has been obtained by reading the document in a bent state.
Abstract: An image forming apparatus includes a reader, a wait part, and a processor. The reader reads an image from a first sheet. In the wait part, the first sheet is made to wait. The processor is configured to make the first sheet wait in the wait part after the reading performed by the reader, and, after the first sheet is made to wait in the wait part, output the first sheet and a second sheet continuously. The second sheet records first information determined on the basis of an evaluation result obtained from evaluation of the first sheet.
Abstract: An image reading device includes a transparent member unit including a transparent member, an opposing member opposed to the transparent member, and a reading mechanism. The reading mechanism irradiates the opposing member and a medium being conveyed between the transparent member and the opposing member with light through the transparent member and receive reflected light from the opposing member and the medium. The reading mechanism includes a position adjuster to which the transparent member unit is secured and a guide to guide the transparent member unit in a direction intersecting a direction of conveyance of the medium to detachably attach the transparent member unit to the reading mechanism. The position adjuster changes a position at which the transparent member unit is secured with respect to the reading mechanism. The transparent member unit is secured to the position adjuster.