Image Segmentation Patents (Class 382/173)
-
Patent number: 10497122Abstract: Various embodiments describe using a neural network to evaluate image crops in substantially real-time. In an example, a computer system performs unsupervised training of a first neural network based on unannotated image crops, followed by a supervised training of the first neural network based on annotated image crops. Once this first neural network is trained, the computer system inputs image crops generated from images to this trained network and receives composition scores therefrom. The computer system performs supervised training of a second neural network based on the images and the composition scores.Type: GrantFiled: October 11, 2017Date of Patent: December 3, 2019Assignee: Adobe Inc.Inventors: Jianming Zhang, Zijun Wei, Zhe Lin, Xiaohui Shen, Radomir Mech
-
Patent number: 10497100Abstract: A video processing system includes a computing platform having a hardware processor and a system memory storing an image cancellation software code. The hardware processor executes the image cancellation software code to receive a frame of video, detect an object image for cancellation from the received frame, and map the received frame from an original representation to a representation in which the object image does not intersect a frame boundary. The image cancellation software code also filters the mapped frame to remove features of the object image that appear to be in motion, inpaint the mapped and filtered frame to mask the object image, and reverse map the mapped and filtered frame having the object image masked to the original representation. The reverse mapped frame is composited with the received frame to produce an inpainted frame of video from which the object image has been cancelled.Type: GrantFiled: October 27, 2017Date of Patent: December 3, 2019Assignee: Disney Enterprises, Inc.Inventors: Christopher Schroers, Alexander Sorkine Hornung
-
Patent number: 10489672Abstract: A data capture component of a mobile device receives information for an identification of a data field in a physical document. The data capture component receives a video stream comprising a plurality of frames, wherein each frame comprises a portion of the physical document. A frame is selected from the plurality of frames in the video stream. One or more text regions in the frame are identified. Each of the identified text region(s) in the frame is processed to identify data of each of the identified text region(s) and to select data of one of the identified text region(s) that corresponds to a set of attributes associated with the data field. The selected data is then compared with data of text regions of a subsequent frame. If the data of the text regions of the subsequent frame is a closer match to the set of attributes, the selected data is updated. A display field is then provided with the selected data for presentation with the frame in a user interface.Type: GrantFiled: June 19, 2017Date of Patent: November 26, 2019Assignee: ABBYY PRODUCTION LLCInventor: Andrey Isaev
-
Patent number: 10484663Abstract: The disclosure relates to an information processing apparatus and information processing method. The information processing apparatus according to an embodiment includes a processing circuitry configured to: derive a disparity map of an object based on at least two images among multi-view images of the object; estimate confidences of disparity values in the disparity map using a classifier trained in advance; and perform an optimization process on the disparity map based on disparity values of reference pixels having confidences higher than a predetermined level.Type: GrantFiled: December 14, 2017Date of Patent: November 19, 2019Assignee: SONY CORPORATIONInventors: Zhaoxin Li, Chongyu Chen, Lei Zhang
-
Patent number: 10467465Abstract: Computerized techniques for improved binarization and extraction of information from digital image data are disclosed in accordance with various embodiments. The inventive concepts include: rendering, using a processor of the mobile device, a digital image using a plurality of binarization thresholds to generate a plurality of range-binarized digital images, wherein each rendering of the digital image is generated using a different combination of the plurality of binarization thresholds; identifying, using the processor of the mobile device, one or more range connected components within the plurality of range-binarized digital images; and identifying, using the processor of the mobile device, a plurality of text regions within the digital image based on some or all of the range connected components. Corresponding systems and computer program products are also disclosed.Type: GrantFiled: December 30, 2016Date of Patent: November 5, 2019Assignee: KOFAX, INC.Inventors: Alexander Shustorovich, Christopher W. Thrasher
-
Patent number: 10467494Abstract: This disclosure relates generally to character detection and recognition, and more particularly to a method and system for container code recognition via Spatial Transformer Networks and Connected Component. The method comprises capturing an image of a container using an image capture device which is pre-processed using an image preprocessing module. The method further comprises extracting and filtering region proposals from the pre-processed image using a region extraction module to generate regrouped region proposals.Type: GrantFiled: December 12, 2017Date of Patent: November 5, 2019Assignee: Tata Consultancy Services LimitedInventors: Monika Sharma, Lovekesh Vig, Ramya Sugnana Murthy Hebbalaguppe, Ehtesham Hassan, Ankit Verma
-
Patent number: 10467744Abstract: Systems and methods for automatically determining of garment sizing (e.g., fit) using a images including video images. The method may perform non-contact estimations of garment fit from visual (e.g., video) input by receiving an video of the subject's head and face and profile and determining a scaling factor from the subject's intraocular spacing and using this scaling factor when analyzing images of other body regions to determine garment sizing.Type: GrantFiled: July 6, 2016Date of Patent: November 5, 2019Assignee: L.I.F.E. Corporation S.A.Inventors: Andrea Aliverti, Gianluigi Longinotti-Buitoni
-
Patent number: 10467504Abstract: Systems, methods, and storage media for evaluating images are disclosed.Type: GrantFiled: February 8, 2019Date of Patent: November 5, 2019Assignee: ADHARK, INC.Inventors: Jehan Hamedi, Zachary Halloran, Elham Saraee
-
Patent number: 10460209Abstract: The present disclosure relates to a vehicle recognition method and system. The vehicle recognition method comprises: receiving a first image; extracting a first vehicle brand recognition region and a first color recognition region of the first image; recognizing the extracted first vehicle brand recognition region by a trained vehicle brand deep learning model; recognizing the extracted first color recognition region; and recognizing the vehicle based on the recognition results of the first vehicle brand recognition region and the first color recognition region.Type: GrantFiled: November 22, 2017Date of Patent: October 29, 2019Assignee: Beijing ICETech Science & Technology Co., Ltd.Inventors: Zhiguo Li, Zhenjie Yang, Lei Geng, Xuebin Wang, Ming Zhu
-
Patent number: 10459726Abstract: Described herein is a system and method for store fusion that fuses small store operations into fewer, larger store operations. The system detects that a pair of adjacent operations are consecutive store operations, where the adjacent micro-operations refers to micro-operations flowing through adjacent dispatch slots and the consecutive store micro-operations refers to both of the adjacent micro-operations being store micro-operations. The consecutive store operations are then reviewed to determine if the data sizes are the same and if the store operation addresses are consecutive. The two store operations are then fused together to form one store operation with twice the data size and one store data HI operation.Type: GrantFiled: November 27, 2017Date of Patent: October 29, 2019Assignee: ADVANCED MICRO DEVICES, INC.Inventor: John M. King
-
Patent number: 10456113Abstract: An ultrasound imaging system includes an image processor and a velocity processor configured to process beamformed ultrasound data representing structure flowing through a tubular object and generate, respectively, an image indicative of the tubular object and vector flow imaging data indicative of the structure flowing through the tubular object. The system further includes a segmentation processor configured to segment the tubular object from the image based on a combination of both the vector flow imaging data and the image, wherein a resulting segmentation extends from wall-to-wall of the tubular object. The system further includes a display configured to display the image with the segmentation and the vector flow imaging data superimposed thereover, with the vector flow imaging data extending from wall-to-wall within the tubular object.Type: GrantFiled: September 8, 2016Date of Patent: October 29, 2019Assignee: B-K Medical ApsInventors: Ramin Moshavegh, Jorgen Arendt Jensen, Bo Martins
-
Patent number: 10462414Abstract: Management of graphical overlays for use in stereoscopic video where the graphical overlay is positioned relative to depths associated with objects showing in the video. A disparity offset processor may be configured to facilitate generating overlay mapping information as a function of object depth maps in order to facilitate positioning one or move graphical overlays relative to corresponding objects.Type: GrantFiled: March 15, 2013Date of Patent: October 29, 2019Assignee: Cable Television Laboratories, Inc.Inventors: David K. Broberg, So Vang, Arianne T. Hinds
-
Patent number: 10458923Abstract: A method for processing print data includes a matching processing step of performing pattern matching between a streak detection pattern including a streak pattern having a width of one pixel and print data after a RIP process, a length measurement step of determining, when matching is established in the matching processing step, a length of a streak candidate part including a part corresponding to the streak pattern in a region where matching is established and a part continuous in an extending direction of the streak pattern and having a same value as a data value of the streak pattern, in the print data, and a determination step of determining whether or not the streak candidate part is white streak data that possibly results in a streak, by comparing the length determined in the length measurement step against a predetermined threshold.Type: GrantFiled: January 12, 2018Date of Patent: October 29, 2019Assignee: SCREEN HOLDINGS CO., LTD.Inventors: Yohei Onuki, Kenichi Yokouchi, Hiroyuki Segawa
-
Patent number: 10460198Abstract: An image processing system comprises a template matching engine (TME). The TME reads an image from the memory; and as each pixel of the image is being read, calculates a respective feature value of a plurality of feature maps as a function of the pixel value. A pre-filter is responsive to a current pixel location comprising a node within a limited detector cascade to be applied to a window within the image to: compare a feature value from a selected one of the plurality of feature maps corresponding to the pixel location to a threshold value; and responsive to pixels for all nodes within a limited detector cascade to be applied to the window having been read, determine a score for the window. A classifier, responsive to the pre-filter indicating that a score for a window is below a window threshold, does not apply a longer detector cascade to the window before indicating that the window does not comprise an object to be detected.Type: GrantFiled: December 15, 2016Date of Patent: October 29, 2019Assignee: FotoNation LimitedInventors: Nicolae Nicoara, Cristina Raceala, Corneliu Zaharia, Szabolcs Fulop, Oana Iovita
-
Patent number: 10452893Abstract: Method, terminal, and storage medium for tracking facial critical area are provided. The method includes accessing a frame of image in a video file; obtaining coordinate frame data of a facial part in the image; determining initial coordinate frame data of a critical area in the facial part according to the coordinate frame data of the facial part; obtaining coordinate frame data of the critical area according to the initial coordinate frame data of the critical area in the facial part; accessing an adjacent next frame of image in the video file; obtaining initial coordinate frame data of the critical area in the facial part for the adjacent next frame of image by using the coordinate frame data of the critical area in the frame; and obtaining coordinate frame data of the critical area for the adjacent next frame of image according to the initial coordinate frame data thereof.Type: GrantFiled: September 26, 2017Date of Patent: October 22, 2019Assignee: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITEDInventor: Chengjie Wang
-
Patent number: 10452903Abstract: An information extraction device includes a storage unit, an image reading device and a control unit. Based on the color of a colored region in a reference original document (reference image data), the control unit classifies the colored region into a first colored region and a second colored region. The storage unit stores feature image data including the first colored region and position data indicating a distance and a direction from the feature image data to the second colored region. The control unit determines an extraction region based on the position of a region matching with the feature image data in comparative image data and the position data. The control unit stores, in the storage unit, extraction information extracted from the extraction region.Type: GrantFiled: July 14, 2017Date of Patent: October 22, 2019Assignee: KYOCERA Document Solutions Inc.Inventor: Atsushi Yamada
-
Patent number: 10444954Abstract: A method and apparatus for rendering geographic areas involves presenting at least part of the geographic area in a distinctive fashion. An area of interest is identified. A location of the area of interest as a geographic sub-area within a geographic area is determined. A representation of the area of interest within an electronic model of the geographic area is located. A view of the electronic model of the geographic area comprising the representation of the area of interest is selected, and the view of the electronic model with the area of interest having a different display characteristic than other geographic sub-areas shown in the view is presented. The different display characteristic distinguishes the area of interest from the other geographic sub-areas.Type: GrantFiled: November 11, 2013Date of Patent: October 15, 2019Assignee: HERE Global B.V.Inventors: Xi Zhang, Xin Chen
-
Patent number: 10445582Abstract: A method of determining a composite action from a video clip, using a conditional random field (CRF), the method includes determining a plurality of features from the video clip, each of the features having a corresponding temporal segment from the video clip. The method may continue by determining, for each of the temporal segments corresponding to one of the features, an initial estimate of an action unit label from a corresponding unary potential function, the corresponding unary potential function having as ordered input the plurality of features from a current temporal segment and at least one other of the temporal segments. The method may further include determining the composite action by jointly optimizing the initial estimate of the action unit labels.Type: GrantFiled: December 20, 2016Date of Patent: October 15, 2019Assignee: Canon Kabushiki KaishaInventor: Nagita Mehrseresht
-
Patent number: 10446113Abstract: A method for inverting raster image-based maps. The method includes obtaining an original raster image of a map and making a determination that nighttime conditions apply. Based on the determination, a processed raster image of the map is obtained by obtaining an intermediate representation in which the hue of the original raster image is inverted and by inverting the hue and tone of the intermediate representation. Subsequently, the processed raster image is displayed.Type: GrantFiled: January 30, 2018Date of Patent: October 15, 2019Assignee: ForeFlight LLCInventor: Joshua Aron Tahmasebi
-
Patent number: 10438085Abstract: One or more pieces of product information each having a product identifier and specifying image information are stored in a product information storage unit. A product identifier acquiring unit detects, at least two images that are to be analyzed, that there are images of products corresponding to one or more pieces of product specifying image information, using the stored pieces of product specifying image information, and acquires one or more product identifiers corresponding to the images of the products. A personal attribute value acquiring unit acquires one or more personal attribute values from the least two images that are to be analyzed, and a personal attribute value-related information accumulating unit accumulates the one or more product identifiers acquired by the product identifier acquiring unit and personal attribute value-related information, in association with each other, the product identifiers and the personal attribute values being information acquired from a same image.Type: GrantFiled: September 13, 2016Date of Patent: October 8, 2019Inventors: Tsz Tat Chu, Mayuko Yamaura
-
Patent number: 10438361Abstract: An image processing apparatus comprises a video input unit, a region division unit configured to divide an image acquired by the video input unit into a plurality of regions each including pixels of similar attributes, a feature extraction unit configured to extract a feature from each divided region, a background model storage unit configured to store a background model generated from a feature of a background in advance, and a feature comparison unit configured to compare the extracted feature with a feature in the background model and determine for each of the regions whether the region is the background.Type: GrantFiled: May 26, 2017Date of Patent: October 8, 2019Assignee: CANON KABUSHIKI KAISHAInventors: Hiroshi Tojo, Kotaro Yano, Tuan quang Pham
-
Patent number: 10430661Abstract: Techniques and systems are described to generate a compact video feature representation for sequences of frames in a video. In one example, values of features are extracted from each frame of a plurality of frames of a video using machine learning, e.g., through use of a convolutional neural network. A video feature representation is generated of temporal order dynamics of the video, e.g., through use of a recurrent neural network. For example, a maximum value is maintained of each feature of the plurality of features that has been reached for the plurality of frames in the video. A timestamp is also maintained as indicative of when the maximum value is reached for each feature of the plurality of features. The video feature representation is then output as a basis to determine similarity of the video with at least one other video based on the video feature representation.Type: GrantFiled: December 20, 2016Date of Patent: October 1, 2019Assignee: Adobe Inc.Inventors: Hao Hu, Zhaowen Wang, Joon-Young Lee, Zhe Lin
-
Patent number: 10410386Abstract: A method for processing an image that includes a table is provided. The method includes: obtaining a set of edges associated with a cell candidate of the table; determining a longest edge in the set of edges and a first orientation of the longest edge; determining, in the set of edges, an initial edge for a first cell border comprising the first orientation and the longest edge; building the first cell border based on the initial edge, the first orientation, and the set of edges; calculating a second orientation for a second cell border by rotating the first orientation; building the second cell border based on an initial edge for the second cell border and the second orientation; and validating the cell candidate as a cell of the table in response to building the first cell border and building the second cell border.Type: GrantFiled: September 15, 2017Date of Patent: September 10, 2019Assignee: KONICA MINOLTA LABORATORY U.S.A., INC.Inventor: Darrell Eugene Bellert
-
Patent number: 10409885Abstract: A device configured to emulate a distance measuring device in a correlithm object processing system. The device is configured to receive an input binary string and to mask a first portion of the input binary string. The device is configured to identify a first binary string in a counting table matching a second portion of the input binary string, to fetch a first numeric value linked with the first binary string, and to increment a count value of a counter by the first numeric value. The device is configured to mask the second portion and to unmask the first portion. The device is configured to identify a second binary string in the counting table matching the first portion, to fetch a second numeric value linked with the second binary string, to increment the count value of the counter by the second numeric value, and to output the count value.Type: GrantFiled: September 11, 2017Date of Patent: September 10, 2019Assignee: Bank of America CorporationInventor: Patrick N. Lawrence
-
Patent number: 10405834Abstract: An anatomical structure is segmented from images generated in an anatomical scan. A user input device is configured for manually identifying two or more seed points on a plurality of the images. An image processor is configured to identify an optimal path through seed points on the plurality of images with a graph search. The optimal path between corresponding seed points on different images defines a boundary contour, and the anatomical structure includes two or more boundary contour. The image processor is further configured to interpolate the boundary contours over a three-dimensional volume using cardinal splines.Type: GrantFiled: May 31, 2016Date of Patent: September 10, 2019Assignee: The Governors of the University of AlbertaInventors: Jacob Lester Jaremko, Abhilash Hareendranathan, Myles Mabee, Richard Thompson, Pierre Boulanger, Kumaradevan Punithakumar
-
Patent number: 10402974Abstract: In a method and an evaluation computer for evaluating medical data having a temporal resolution, a preprocessing phase and a postprocessing phase, are executed. The medical data include a first dataset acquired at a first time point of the temporal resolution and a second dataset acquired at a second time point of the temporal resolution. The preprocessing of the first dataset is performed in a first time period and the preprocessed first dataset is provided as an output. The preprocessing of the second dataset is performed in a second time period and the preprocessed second dataset is provided as an output, and an interactive preparation of the postprocessing is carried out in the second time period based on the preprocessed first dataset. The medical data are evaluated by postprocessing using the preprocessed first dataset and/or the preprocessed second dataset.Type: GrantFiled: September 29, 2017Date of Patent: September 3, 2019Assignee: Siemens Healthcare GmbHInventor: Thomas Beck
-
Patent number: 10402689Abstract: A machine learning system can generate an image mask (e.g., a pixel mask) comprising pixel assignments for pixels. The pixels can be assigned to classes, including, for example, face, clothes, body skin, or hair. The machine learning system can be implemented using a convolutional neural network that is configured to execute efficiently on computing devices having limited resources, such as mobile phones. The pixel mask can be used to more accurately display video effects interacting with a user or subject depicted in the image.Type: GrantFiled: September 15, 2017Date of Patent: September 3, 2019Assignee: Snap Inc.Inventors: Lidiia Bogdanovych, William Brendel, Samuel Edward Hare, Fedir Poliakov, Guohui Wang, Xuehan Xiong, Jianchao Yang, Linjie Yang
-
Patent number: 10402751Abstract: A method includes performing, by a processor: receiving a document containing subject matter related to a course of action, the document comprising a plurality of sub-documents that are related to one another in a time sequence, converting the document to a vector format to generate a vectorized document that encodes a probability distribution of words in the document and transition probabilities between words, applying a machine learning algorithm to the vectorized document to generate an estimated vectorized document, associating the estimated vectorized document with a reference document; predicting future subject matter contained in a future sub-document of the document based on the reference document, and adjusting the course of action responsive to predicting the future subject matter.Type: GrantFiled: March 21, 2016Date of Patent: September 3, 2019Assignee: CA, Inc.Inventors: Jaume Ferrarons Llagostera, David Sánchez Charles, Victor Muntés Mulero
-
Patent number: 10402976Abstract: A framework for isolating an aneurysm and parent vessel in volumetric image data is provided herein. In accordance with one aspect, the framework generates a refined mask by performing region growing starting at an aneurysm dome point to eliminate vessels that are indirectly connected to an aneurysm or parent vessel in volumetric image data. A final mask may be generated based at least in part on the refined mask by eliminating any kissing vessel connected with the aneurysm from the refined mask. The final mask may then be used for segmentation of the aneurysm and the parent vessel in the volumetric image data.Type: GrantFiled: May 16, 2017Date of Patent: September 3, 2019Assignee: Siemens Healthcare GmbHInventors: Komal Dutta, Scott Baker
-
Patent number: 10404978Abstract: Disclosed is a driving method and a driving system for reducing a residual image of AMOLED display. The driving method includes: performing a high contrast edge detection on a static picture; determining an edge contrast of each pixel based on a result of the edge detection; adjusting an output luminance proportional coefficient of each pixel based on the edge contrast of each pixel and a position of each pixel itself; and outputting a luminance of each pixel by the output luminance proportional coefficient corresponding to each pixel. The driving method can improve the bright and dark difference ratio of the edge contours in the high contrast region of the image.Type: GrantFiled: December 22, 2016Date of Patent: September 3, 2019Assignee: SHENZHEN CHINA STAR OPTOELECTRONICS TECHNOLOGY CO., LTD.Inventors: Yuchao Zeng, Tai Jiun Hwang
-
Patent number: 10395094Abstract: This application discloses a method and a terminal for detecting glasses in a face image. The method includes: obtaining a face image; determining a nose bridge region in the face image; detecting an image change in the nose bridge region to obtain an image change result of the nose bridge region; and determining whether there are glasses in the face image according to the image change result of the nose bridge region. The terminal for detecting glasses in a face image matches the method.Type: GrantFiled: September 5, 2017Date of Patent: August 27, 2019Assignee: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITEDInventors: Chengjie Wang, Guofu Tan, Hui Ni
-
Patent number: 10388017Abstract: The present invention provides a method for detection of different ontologies using advanced unsupervised machine learning which will be used to visualize factors not visible to the human observer, such as unknown characteristics between imaging datasets and other factors to provide insights into the structure of the data. This methodology is referred to herein as Contribution Scattergram. An example includes using radiological images to determine a relationship in dimension, structure, and distance between each parameter. This information can be used to determine if changes in the images have occurred and for treatment response.Type: GrantFiled: July 31, 2014Date of Patent: August 20, 2019Assignee: The Johns Hopkins UniversityInventors: Michael A. Jacobs, Alireza Akhbardeh
-
Patent number: 10389952Abstract: An image-processing device includes a high-resolution combining portion that generates a combined image by combining a standard image and at least one reference image, which are acquired by capturing images of a subject in time series by using an imaging element in which multiple types of color filters are arrayed in individual pixels, in a high-resolution space where the resolution is greater than those of the standard image and the reference images.Type: GrantFiled: November 21, 2017Date of Patent: August 20, 2019Assignee: OLYMPUS CORPORATIONInventors: Eiji Furukawa, Kosuke Kajimura
-
Patent number: 10380723Abstract: In some embodiments, an image editing application stores, based on a first selection input, a selection state that identifies a first image portion of a target image as included in a preview image displayed in a mask-based editing interface of the image editing application. An edit to the preview image generated from the selected first image portion is applied in the mask-based editing interface. The image editing application also updates an edit state that tracks the edit applied to the preview image. The image editing application modifies, based on a second selection input received via the mask-based editing interface, the selection state to include a second image portion in the preview image. The edit state is maintained with the applied edit concurrently with modifying the selection state. The image editing application applies the edit to the modified preview image in the mask-based editing interface.Type: GrantFiled: June 19, 2017Date of Patent: August 13, 2019Assignee: Adobe Inc.Inventors: Betty M. Leong, Alan L. Erickson, Sarah Stuckey, Sarah Aye Kong, Bradee R. Evans
-
Patent number: 10359994Abstract: A computer system processing heterogeneous measurements from various metrology apparatuses includes a processor, a storage, and a software platform. The storage includes a recording of direct models expressing the heterogeneous measurements as a function of predefined features of electronic devices or parameters for processing the measurements. The software platform includes a functional cost calculation module and an interface selecting one of plural cost functions. The functional module provides an estimation of a cost by comparison of the heterogeneous measurements with an estimation of the measurements obtained by an application of the direct models to values of predefined features or processing parameters. The software platform includes a functional solver module and an interface selecting one of plural solvers, for an iterative optimization by the solver of the values based on the output of the cost calculation module and by inversion of the direct models.Type: GrantFiled: November 3, 2014Date of Patent: July 23, 2019Assignee: Commissariat a l'energie atomique et aux energies alternativesInventors: Jerome Hazart, Johann Foucher
-
Patent number: 10348958Abstract: Predetermined processing is reliably performed on an image captured when a person serving as a photographic subject is photographed. When a person's face is detected in an image captured by an imaging section, a control section performs predetermined image processing on the captured image. If a person's face is not detected, the control section judges whether the predetermined processing is required to be performed on the captured image and, when judged that the predetermined processing is required to be performed, gives an instruction to perform the predetermine processing on the captured image.Type: GrantFiled: June 3, 2016Date of Patent: July 9, 2019Assignee: CASIO COMPUTER CO., LTD.Inventor: Takeshi Sato
-
Patent number: 10339595Abstract: A system and method for computer vision driven applications in an environment that can include collecting image data across an environment; maintaining an environmental object graph from the image data whereby maintaining the environmental object graph is an iterative process that includes: classifying objects, tracking object locations, detecting interaction events, instantiating object associations in the environmental object graph, and updating the environmental object graph by propagating change in at least one object instance across object associations; and inspecting object state for at least one object instance in the environmental object graph and executing an action associated with the object state. The system and method can be applied to automatic checkout, inventory management, and/or other system integrations.Type: GrantFiled: May 9, 2017Date of Patent: July 2, 2019Assignee: Grabango Co.Inventors: William Glaser, Brian Van Osdol
-
Patent number: 10332261Abstract: A method may include identifying a first image for training a deep learning network, wherein the first image includes at least one target object associated with at least one location in the first image, and wherein the first image is associated with a mask image; determining a set of deformations to create a training set of deformed images, wherein the training set is to be used to train the deep learning network; generating the training set of deformed images by applying the set of deformations to the first image; and generating a set of deformed mask images by applying the set of deformations to the mask image, wherein each deformed image of the training set of deformed images is associated with a respective mask image to identify the location of the at least one target object in each deformed image.Type: GrantFiled: April 26, 2018Date of Patent: June 25, 2019Assignee: Capital One Services, LLCInventors: Reza Farivar, Kenneth Taylor
-
Patent number: 10326780Abstract: Systems and methods for presenting information associated with an application are disclosed. An application is executed at a computing device, wherein the application presents a visual output at a display unit of the computing device. A guest process is run within the application. The guest process stores and processes untrusted content. An embedder process is run within the application and in parallel with the guest process. The embedder process stores and processes trusted content. The application receives, via the guest process, a request to present an image associated with the untrusted content. The embedder process specifies size constraints for the image. The guest process adjusts the image to fit within the size constraints specified via the embedder process.Type: GrantFiled: September 6, 2016Date of Patent: June 18, 2019Assignee: GOOGLE LLCInventor: Fady Samuel
-
Patent number: 10319096Abstract: In some implementations, a computer-implemented method is capable of automatically segmenting and detecting a tattoo within an image. An image may be initially obtained. A block coverage pattern that identifies multiple blocks within the obtained image may be determined. A set of processing operations may then be performed for each block. The processing operations may include calculating a plurality of statistical features. A confidence score reflecting a likelihood that at least a portion of the block includes a predetermined graphical attribute associated with tattoos may be calculated. A subset of the multiple blocks of the image that have a respective confidence score greater than a predetermined threshold value may be identified. A portion of the image that includes one or more blocks from among the subset of the multiple blocks may then be determined to correspond to a tattoo.Type: GrantFiled: December 18, 2018Date of Patent: June 11, 2019Assignee: MorphoTrak, LLCInventors: Darrell Hougen, Peter Zhen-Ping Lo
-
Patent number: 10319115Abstract: Provided is an image compression device including an object extracting unit configured to perform convolution neural network (CNN) training and identify an object from an image received externally, a parameter adjusting unit configured to adjust a quantization parameter of a region in which the identified object is included in the image on the basis of the identified object, and an image compression unit configured to compress the image on the basis of the adjusted quantization parameter.Type: GrantFiled: September 7, 2017Date of Patent: June 11, 2019Assignee: ELECTRONICS & TELECOMMUNICATIONS RESEARCH INSTITUTEInventors: Seong Mo Park, Sung Eun Kim, Ju-Yeob Kim, Jin Kyu Kim, Kwang Il Oh, Joo Hyun Lee
-
Patent number: 10311331Abstract: An image processing apparatus includes a feature value extractor and an image adjuster. The feature value extractor extracts feature values of a first specified image, which is an image of a region specified in advance in a first image. The image adjuster changes, in a case where the feature values of the first specified image are to be reflected in a second image, for each region of the second image, a degree to which the feature values of the first specified image are to be reflected, and adjusts an image quality of the second image.Type: GrantFiled: August 25, 2016Date of Patent: June 4, 2019Assignee: FUJI XEROX CO., LTD.Inventors: Kosuke Maruyama, Keiko Matsubayashi, Makoto Sasaki, Takayuki Yamamoto
-
Patent number: 10304192Abstract: The two-step architecture of the supervoxel-based spatial temporal video segmentation method ensures speed and scalability. The computationally intensive first step uses a highly efficient super-voxel segmentation method. The second step is done over pre-grouped super-voxels, hence has much lower temporal and spatial complexity. The progressive segmentation scheme deployed in the first step enables segmenting huge input volume part by part, without loading all the data into the memory which may be infeasible. At the same time, the progressive segmentation is able to effectively prevent seam artifacts, leading to segmentation results virtually identical to those of whole volume processing.Type: GrantFiled: July 11, 2017Date of Patent: May 28, 2019Assignee: Sony CorporationInventors: Xun Xu, Akira Nakamura
-
Patent number: 10303971Abstract: A method of classifying an image is disclosed. The method comprises: applying a computer vision procedure to the image to detect therein candidate image regions suspected as being occupied by a target; presenting to an observer each candidate image region as a visual stimulus, while collecting neurophysiological signals from a brain of the observer; processing the neurophysiological signals to identify a neurophysiological event indicative of a detection of the target by the observer; and determining an existence of the target in the image is based, at least in part, on the identification of the neurophysiological event.Type: GrantFiled: June 2, 2016Date of Patent: May 28, 2019Assignee: InnerEye Ltd.Inventors: Amir B. Geva, Leon Y. Deouell, Sergey Vaisman, Omri Harish, Ran El Manor, Eitan Netzer, Shani Shalgi
-
Patent number: 10296535Abstract: According to one embodiment, a content item is received that was identified in response to a search query received from a client. It is determined whether the search query is within a predetermined percentage of total search queries that yield the content item. In response to determining that the search query is within the predetermined percentage of the total search queries, a search is performed in an image store based on one or more keywords associated with the search query to identify a list of one or more images and one of the identified images is selected to be matched with the content item. Otherwise, a predetermined preferred image is selected to be matched with the content item. The content item is incorporated with the selected image to generate a composite content item to be transmitted to the client.Type: GrantFiled: August 23, 2016Date of Patent: May 21, 2019Assignee: BAIDU USA LLCInventors: Dong Li, Yifei Jiang, Zhuang Li, Fan Zhu, Qi Kong, Guang Yang, Jingao Wang
-
Patent number: 10296933Abstract: Particular embodiments may access an electronic image including a surface area and divide the electronic image into a plurality of surfaces. Particular embodiments may then determine that one or more of the surfaces include a type of graphics and determine a percentage of the surface area of the image that is occupied by the one or more surfaces determined to include the type of graphics.Type: GrantFiled: April 12, 2013Date of Patent: May 21, 2019Assignee: Facebook, Inc.Inventors: Frederick Gottesman, David Clune, James Andrews, Igor Gevka, Satwik Shukla
-
Patent number: 10289884Abstract: According to an embodiment, an image analyzer includes a set unit, a first calculator, a second calculator, a divider, a third calculator, and a specifier. The set unit sets multiple partial regions in the image. The first calculator calculates an optical flow in a partial region. The second calculator calculates the density of an object included in the partial region. The divider divides the image into the multiple blocks. The third calculator calculates, for a block, a statistical amount that is derived from the density and the moving amount of the object derived from the optical flow for the partial region included in the block. The specifier specifies the block, out of the blocks, as an unsteady block that is in an unsteady state on the basis of the statistical amount.Type: GrantFiled: August 25, 2016Date of Patent: May 14, 2019Assignee: Kabushiki Kaisha ToshibaInventors: Quoc Viet Pham, Tomoki Watanabe
-
Patent number: 10290086Abstract: One or more embodiments of an image enhancement system enable a computing device to generate an enhanced digital image. In particular, a computing device can enhance a digital image including, for example, a photograph of a whiteboard, document, chalkboard, or other object having a uniform background. The computing device can determine modifications to apply to the digital image by minimizing an energy heuristic that both causes pixels of the digital image to change to a uniform color (e.g., white) and preserves gradients from the digital image. The computing device can further generate an enhanced digital image by applying the determined modifications to the digital image.Type: GrantFiled: October 2, 2017Date of Patent: May 14, 2019Assignee: DROPBOX, INC.Inventor: Jongmin Baek
-
Patent number: 10282420Abstract: A method, an apparatus and a system for recognizing an evaluation element are provided. The method includes receiving an input text; performing, using a first conditional random field model, first recognition for the input text to obtain a first recognition result, the first recognition result including a pre-evaluation element that is recognized by using the first conditional random field model; performing, using a second conditional random field model, second recognition for the input text to obtain a second recognition result, the second recognition result including a false positive evaluation element that is recognized by using the second conditional random field model, the false positive evaluation element being an element erroneously detected as an evaluation element; and recognizing, based on the first recognition result and the second recognition result, an evaluation element in the input text.Type: GrantFiled: May 17, 2017Date of Patent: May 7, 2019Assignee: Ricoh Company, Ltd.Inventors: Shanshan Jiang, Bin Dong, Jichuan Zheng, Jiashi Zhang, Yixuan Tong
-
Patent number: 10278661Abstract: A dynamic diagnosis support information generation system includes: a radiation generator capable of irradiating a pulsed radiation; a radiation detector which is provided with a plurality of detecting elements arranged in two-dimension, detects the pulsed radiation irradiated from the radiation generator at each of the plurality of detecting elements and generates frame images successively; and an analysis section which calculates and outputs a feature value relating to a dynamic image of a subject based on a plurality of frame images generated by radiographing the subject by using the radiation generator and the radiation detector, wherein the analysis section calculates the feature value relating to the dynamic image of the subject by corresponding pixels to each others representing outputs of a detecting element at a same position in the radiation detector among the plurality of the frame images.Type: GrantFiled: November 17, 2014Date of Patent: May 7, 2019Assignee: KONICA MINOLTA, INC.Inventors: Shintaro Muraoka, Tetsuo Shimada, Sho Noji