Abstract: A system, according to various embodiments, receives images such as photographs and videos from one or more portable computing devices associated with one or more individuals (e.g., construction workers or landscapers) while the portable computing devices are in a particular position within a particular location at a particular time. The system determines a virtual position within a 3-D representation of the particular location that generally corresponds to the particular position and combines the images with the 3-D representation to generate an enhanced 3-D representation of the particular location. This may allow, for example, owners of a particular property to track and quickly understand construction and landscaping work that has been done on their property and to easily contact those individuals regarding that work.
Abstract: Described is a system for controlling autonomous platform. Based on an input image, the system generates a motor control command decision for the autonomous platform. A probability of the input image belonging to a set of training images is determined, and a reliability measure for the motor control command decision is generated using the determined probability. An exploratory action is performed when the reliability measure is above a predetermined threshold. Otherwise, an exploitation action corresponding with the motor control command decision is performed when the reliability measure is below a predetermined threshold.
Abstract: Lossless compression of fragmented image data is disclosed. In some embodiments, a stream of information comprising data elements having statistical characteristics is received. An encoded output is produced by an encoder comprising a data compressor that implements a variable length code that is adapted to the statistical characteristics of the data elements. The output and information from which the variable length code can be derived are stored.
Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for determining, by a pixelation device, an arrangement of pixels for generating representations of data for a document; adjusting, by an increment module and based on the arrangement of pixels, an optical density of one or more pixels included in the arrangement; associating, by the pixelation device, a plurality of distinct types of credential data with one or more line patterns included in the arrangement of pixels; and generating, by the pixelation device, multiple line patterns at a document, the multiple line patterns encoding the plurality of distinct types of credential data.
Type:
Grant
Filed:
December 31, 2018
Date of Patent:
July 27, 2021
Assignee:
Idemia Identity & Security USA LLC
Inventors:
Robert L. Jones, Yecheng Wu, Daoshen Bi
Abstract: Embodiments of the disclosure provide methods and systems for acquiring map data. The system may include a mounting structure configured to adjustably mount a sensor to a vehicle. The sensor may be configured to capture data indicative of at least one surrounding object as the vehicle travels along a path. The system may further include a controller configured to dynamically determine a mounting angle based on the captured data, and cause the mounting structure to adjust the sensor according to the dynamically determined mounting angle.
Type:
Grant
Filed:
December 29, 2018
Date of Patent:
July 13, 2021
Assignee:
Beijing DiDi Infinity Technology and Development Co., Ltd.
Abstract: The present disclosure relates to methods and systems for data visualization. The systems may perform the methods to obtain a video having a plurality of frames including a plurality of objects; identify a target object from the plurality of objects according to the plurality of frames; determine one or more track points of the target object, each of the one or more track points being corresponding to the target object in one of the plurality of frames; determine a first track of the target object based on the track points, the first track including at least one of the one or more track points of the target object; determine a second track of the target object based on the first track, the second track including at least one of the track points of the first track; generate a video analysis result by analyzing the second track; and visualize the video analysis result.
Abstract: According to the present invention, switching of the monitoring images matching the intention of the observer can be automatically performed for images from a plurality of image capturing apparatus, and the load about the job of the observer can be reduced. The image monitoring apparatus includes an estimating unit configured to estimate attention degrees of a user for a plurality of images acquired from the plurality of image capturing apparatuses, a designating unit configured to designate one of the acquired images as an image to be displayed in accordance with an instruction from the user, a learning unit configured to cause the estimating unit to learn so as to increase an attention degree of the designated image, and a selecting unit configured to select one of the plurality of images based on an attention degree of each estimated image.
Abstract: Lossless compression of fragmented image data is disclosed. In some embodiments, a stream of information is received, wherein the stream of information comprises a sequence of tuples and wherein each of the tuples comprises data elements corresponding to one of a plurality of input channels. A channel transformer is employed to rearrange the data elements into a plurality of output channels for an output stream wherein the output channels have higher compressibility than the input channels. The compressed output stream is stored.
Abstract: A method of identifying a living creature includes training a convolutional neural network model using pretrained convolutional neural networks to generate proposals about the regions where there might be an anatomical object within a digital image. Introducing a residual connection to get the input from the previous layer to the next layer helps in solving gradient vanishing problem. The next step is to design an object detector network that does three tasks: classifying the boxes with respective anatomies, tightening the boxes, and generating a mask (i.e., pixel-wise segmentation) of each anatomical component. In constructing the architecture of the object detector network, the network uses per-pixel sigmoid, and binary cross-entropy loss function (to identify the k anatomical components) and rigorously train them.
Abstract: Embodiments of the present specification provide a method and system for improving the quality of a vehicle damage image on the basis of a GAN network. During operation, the system obtains a first vehicle damage image and inputs the first vehicle damage image to a machine-learning model to obtain a second vehicle damage image. The machine-learning model is trained using a plurality of labeled samples of vehicle damage images, and the second vehicle damage image has a better quality than the first vehicle damage image.
Abstract: Implementations of the present disclosure can include a method and apparatus for processing point cloud data. Specifically, the method for processing point cloud data can be provided, including: acquiring a first frame and a second frame respectively from the point cloud data; extracting a first candidate object from the first frame and a second candidate object corresponding to the first candidate object from the second frame, respectively; determining a first location of the first candidate object and a second location of the second candidate object in a coordinate system of the point cloud data, respectively; and identifying any one of the first candidate object and the second candidate object as a moving object, in response to an offset between the first location and the second location.
Abstract: Supervised machine learning using neural networks is applied to denoising images rendered by MC path tracing. Specialization of neural networks may be achieved by using a modular design that allows reusing trained components in different networks and facilitates easy debugging and incremental building of complex structures. Specialization may also be achieved by using progressive neural networks. In some embodiments, training of a neural-network based denoiser may use importance sampling, where more challenging patches or patches including areas of particular interests within a training dataset are selected with higher probabilities than others. In some other embodiments, generative adversarial networks (GANs) may be used for training a machine-learning based denoiser as an alternative to using pre-defined loss functions.
Type:
Grant
Filed:
February 12, 2020
Date of Patent:
June 15, 2021
Assignees:
Pixar, Disney Enterprises, Inc.
Inventors:
Thijs Vogels, Fabrice Rousselle, Brian McWilliams, Mark Meyer, Jan Novak
Abstract: An image-processing apparatus including: an image processor including circuitry or a hardware processor that operates under control of a stored program, the image processor being configured to execute processes including: a saliency-map calculating process that calculates saliency maps on a basis of at least one type of feature quantity obtained from an input image; a salient-region-identifying process that identifies a salient region by using the saliency maps; a salient-region-score-calculating process that calculates a score of the salient region by comparing a distribution of values of the saliency map in the salient region and a distribution of values of the saliency map in a region other than the salient region; and a saliency-evaluating process that evaluates the saliency of the salient region on a basis of the score.
Abstract: System and method for determining physiological parameters of a person are disclosed. A physiological parameter may be obtained by analyzing a facial image of a person, and determining, from the facial image, a physiological parameter of the person by processing the facial image with a data processor. A neural network model such as regression deep learning convolutional neural network is used to predict the physiological parameter. An image processor screens out images which can't be recognized as facial images and adjust facial images to frontal facial images for predicting of physiological parameters.
Type:
Grant
Filed:
April 5, 2018
Date of Patent:
June 8, 2021
Inventors:
Walter De Brouwer, Apurv Mishra, Samia De Brouwer
Abstract: Systems, apparatuses, and methods may provide for technology to process multi-resolution images by identifying pixels at a boundary between pixels of different resolutions, and selectively smoothing the identified pixels.
Type:
Grant
Filed:
April 16, 2020
Date of Patent:
June 8, 2021
Assignee:
Intel Corporation
Inventors:
Travis T. Schluessler, Joydeep Ray, John H. Feit, Nikos Kaburlasos, Jacek Kwiatkowski
Abstract: The present invention relates to a communication technique, which is a convergence of IoT technology and 5G communication system for supporting higher data transmission rate beyond 4G system, and a system for same. The present invention can be applied to smart services (e.g. smart homes, smart buildings, smart cities, smart cars or connected cars, health care, digital education, retail businesses, security- and safety-related services and the like) on the basis of 5G communication technology and IoT-related technology.
Type:
Grant
Filed:
January 9, 2017
Date of Patent:
May 11, 2021
Assignee:
SAMSUNG ELECTRONICS CO., LTD.
Inventors:
Hae-In Chun, Seong-Hwan Oh, Dae-Eun Yi, Sung-Do Choi, Yang-Wook Kim, In-Hak Na
Abstract: A system, method, and non-transitory computer readable medium are provided for training and applying defect classifiers in wafers having deeply stacked layers. In use, a plurality of images generated by an inspection system for a location of a defect detected on a wafer by the inspection system are acquired. The location on the wafer is comprised of a plurality of stacked layers, and each image of the plurality of images is generated by the inspection system at the location using a different focus setting. Further, a classification of the defect is determined, utilizing the plurality of images.
Abstract: A method for processing image array data is provided, which may include the following steps: providing the image array data detected by a sensor at a first time point and including a plurality of first blocks, and each of the first blocks has a color table corresponding thereto; segmenting the image array data detected by a sensor at a second time point into a plurality of second blocks; comparing the image array data of each first block with the image array data of each second block to generate numerical difference information; and respectively allocating the color tables corresponding to the first blocks to the second blocks matching the first blocks according to the numerical difference information.
Type:
Grant
Filed:
November 26, 2018
Date of Patent:
March 30, 2021
Assignee:
INDUSTRIAL TECHNOLOGY RESEARCH INSTITUTE
Abstract: Systems and methods for predicting pest susceptibility, comprising steps to receive geocoded geospatial image data of a crop field from sensors, receive microclimate data of the crop field, determine a crop vigor map for the crop field, and then generate a pest susceptibility map utilizing a risk model based on the crop vigor map and the microclimate data. The pest susceptibility map comprises a measure of a susceptibility of a crop in the crop field to one or more crop pests at one or more locations. In some embodiments, the method also comprises steps to generate a treatment plan (e.g., pesticide application) and to estimate an anticipated return on investment (ROI). The system therefore leverages remote-sensed data, machine data, analytics, and machine learning to enable farmers to predict, prevent, and control the outbreak of crop pests to greatest economic effect. Such a system addresses a fundamental problem in agriculture.
Abstract: Disclosed is a portable 3-dimensional (3D) document scanning device and method. The portable 3D document scanning device includes a projector projecting light of a predetermined pattern onto a document, a 3D scanning camera capturing a pattern of the light projected onto the document and the document, and a control unit estimating a size of the captured document based on image data of the captured document, estimating a shape of the captured document based on the pattern of the projected light and the captured pattern of the light, and correcting the image data of the captured document based on the estimated size and the estimated shape.