Interframe Coding (e.g., Difference Or Motion Detection) Patents (Class 382/236)
  • Patent number: 11887366
    Abstract: A method comprising performing object detection within a set of representations of a hierarchically-structured signal, the set of representations comprising at least a first representation of the signal at a first level of quality and a second representation of the signal at a second, higher level of quality.
    Type: Grant
    Filed: February 11, 2020
    Date of Patent: January 30, 2024
    Inventors: Guido Meardi, Guendalina Cobianchi, Balázs Keszthelyi, Ivan Makeev, Simone Ferrara, Stergios Poularakis
  • Patent number: 11889227
    Abstract: A method for converting a frame rate of an input video is provided. The method includes performing a motion estimation by generating at least one motion field for at least one pair of frames of the input video, wherein the at least one motion field includes a set of motion vectors for each block of a first reference frame of the input video, the set of motion vectors indicating to a second reference frame of the input video, and preparing first data for a predetermined interpolation phase for each block of an interpolated frame, where the first data include at least one parameter for a pre-trained occlusion processing (OcC) convolutional neural network (CNN), the at least one parameter being obtained based on the at least one motion field.
    Type: Grant
    Filed: November 15, 2021
    Date of Patent: January 30, 2024
    Assignee: SAMSUNG ELECTRONICS CO., LTD.
    Inventors: Petr Pohl, Sergei Dmitrievich Ilichev, Igor Mironovich Kovliga, Kristina Olegovna Rakova
  • Patent number: 11889122
    Abstract: A technique for streaming and a client device that uses the technique are disclosed herein. The disclosed technique determines context complexity of streamed data and determines whether to discard or select the streamed data for a future reference frame based on the context complexity of the streamed data. The streamed data is discarded if the content complexity is higher than a content complexity threshold, and the streamed data is selected if the content complexity is not higher than a content complexity threshold. This is based on the realization that error propagation in the case of a less complex video sequence is not very bothersome to the end user experience whereas corruption will be very severe in cases of highly complex sequences.
    Type: Grant
    Filed: August 23, 2021
    Date of Patent: January 30, 2024
    Assignee: NVIDIA Corporation
    Inventors: Thrinadh Kottana, Vinayak Pore, Chirayu Garg, Soumen Kumar Dey
  • Patent number: 11876982
    Abstract: A method for encoding a video sequence is provided that includes signaling in the compressed bit stream that a subset of a plurality of partitioning modes is used for inter-prediction of a portion of the video sequence, using only the subset of partitioning modes for prediction of the portion of the video sequence, and entropy encoding partitioning mode syntax elements corresponding to the portion of the video sequence, wherein at least one partitioning mode syntax element is binarized according to a pre-determined binarization corresponding to the subset of partitioning modes, wherein the pre-determined binarization differs from a pre-determined binarization for the least one partitioning mode syntax element that would be used if the plurality of partitioning modes is used for inter-prediction.
    Type: Grant
    Filed: September 9, 2020
    Date of Patent: January 16, 2024
    Assignee: Texas Instruments Incorporated
    Inventor: Minhua Zhou
  • Patent number: 11830234
    Abstract: An image processing method and an image processing apparatus are provided. The image processing method includes acquiring information of a first region of interest (ROI) in a first frame, estimating information of a second ROI in a second frame that is received after the first frame, based on the acquired information of the first ROI, and sequentially storing, in a memory, subframes that are a portion of the second frame, each of the subframes being a line of the second frame. The image processing method further includes determining whether a portion of the stored subframes includes the second ROI, based on the estimated information of the second ROI, and based on the portion of the stored subframes being determined to include the second ROI, processing the portion of the stored subframes.
    Type: Grant
    Filed: November 15, 2022
    Date of Patent: November 28, 2023
    Assignee: SAMSUNG ELECTRONICS CO., LTD.
    Inventors: Jingu Heo, Byong Min Kang
  • Patent number: 11776164
    Abstract: Techniques are provided herein for more efficiently storing images that have a common subject, such as product images that share the same product in the image. Each image undergoes an adaptive tiling procedure to split the image into a plurality of tiles, with each tile identifying a region of the image having pixels with the same content. The tiles across multiple images can then be clustered together and those tiles having identical content are removed. Once all duplicate tiles have been removed from the set of all tiles across the images, the tiles are once again clustered based on their encoding scheme and certain encoding parameters. Tiles within each cluster are compressed using the best compression technique for the tiles in each corresponding cluster. By removing duplicative tile content between numerous images of the same subject, the total amount of data that needs to be stored is reduced.
    Type: Grant
    Filed: December 22, 2022
    Date of Patent: October 3, 2023
    Assignee: Adobe, Inc.
    Inventors: Ajay Bedi, Rishu Aggarwal
  • Patent number: 11770550
    Abstract: Provided are an encoding device using predictive coding in a screen and a method thereof, and a decoding device and a method thereof. The encoding device is capable of generating a prediction block of a current block according to one of methods for generating a plurality of predetermined prediction blocks, and outputting at least one of encoding information and differential blocks as bitstream by entropy encoding according to whether differential blocks are encoded or not.
    Type: Grant
    Filed: August 7, 2019
    Date of Patent: September 26, 2023
    Assignees: Electronics and Telecommunications Research Institute, UNIVERSITY-INDUSTRY COOPERATION GROUP OF KYUNG HEE UNIVERSITY
    Inventors: Sung Chang Lim, Hae Chul Choi, Se Yoon Jeong, Suk Hee Cho, Hui Yong Kim, Jin Ho Lee, Ha Hyun Lee, Jong Ho Kim, Jin Soo Choi, Jin Woong Kim, Gwang Hoon Park, Kyung Yong Kim, Dong Gyu Sim, Seoung Jun Oh
  • Patent number: 11692885
    Abstract: The present invention proposes a method for identifying the spatial-temporal distribution of the vehicle loads on a bridge based on the DenseNet. The method includes five steps: firstly, mounting a plurality of cameras in different positions of a bridge, acquiring images of the bridge from different directions, and outputting video images with time tags; secondly, acquiring multichannel characteristics of vehicles on the bridge by using DenseNet, including color characteristics, shape characteristics and position characteristics; thirdly, analyzing the data and characteristics of the vehicles from different cameras at a same moment to obtain vehicle distribution on the bridge at any time; fourthly, continuously monitoring the vehicle distribution in a time period to obtain a vehicle load situation on any section of the bridge; and finally, integrating the time and space distribution of the vehicles to obtain spatial-temporal distribution of the bridge.
    Type: Grant
    Filed: September 3, 2020
    Date of Patent: July 4, 2023
    Assignee: Zhejiang University
    Inventors: Xiaowei Ye, Zhexun Li, Tao Jin
  • Patent number: 11683491
    Abstract: Computer processor hardware receives image data specifying element settings for each image of multiple original images in a sequence. The computer processor hardware analyzes the element settings across the multiple original images. The computer processor hardware then utilizes the element settings of the multiple original images in the sequence to produce first encoded image data specifying a set of common image element settings, the set of common image element settings being a baseline to substantially reproduce each of the original images in the sequence.
    Type: Grant
    Filed: September 3, 2020
    Date of Patent: June 20, 2023
    Inventors: Luca Rossato, Guido Meardi
  • Patent number: 11562508
    Abstract: Techniques are provided herein for more efficiently storing images that have a common subject, such as product images that share the same product in the image. Each image undergoes an adaptive tiling procedure to split the image into a plurality of tiles, with each tile identifying a region of the image having pixels with the same content. The tiles across multiple images can then be clustered together and those tiles having identical content are removed. Once all duplicate tiles have been removed from the set of all tiles across the images, the tiles are once again clustered based on their encoding scheme and certain encoding parameters. Tiles within each cluster are compressed using the best compression technique for the tiles in each corresponding cluster. By removing duplicative tile content between numerous images of the same subject, the total amount of data that needs to be stored is reduced.
    Type: Grant
    Filed: December 13, 2020
    Date of Patent: January 24, 2023
    Assignee: Adobe, Inc.
    Inventors: Ajay Bedi, Rishu Aggarwal
  • Patent number: 11527053
    Abstract: An image processing method and an image processing apparatus are provided. The image processing method includes acquiring information of a first region of interest (ROI) in a first frame, estimating information of a second ROI in a second frame that is received after the first frame, based on the acquired information of the first ROI, and sequentially storing, in a memory, subframes that are a portion of the second frame, each of the subframes being a line of the second frame. The image processing method further includes determining whether a portion of the stored subframes includes the second ROI, based on the estimated information of the second ROI, and based on the portion of the stored subframes being determined to include the second ROI, processing the portion of the stored subframes.
    Type: Grant
    Filed: August 16, 2019
    Date of Patent: December 13, 2022
    Assignee: SAMSUNG ELECTRONICS CO., LTD.
    Inventors: Jingu Heo, Byong Min Kang
  • Patent number: 11523132
    Abstract: A predictive image acquisition device is a predictive image acquisition device for associating an observation signal, the observation signal being a signal obtained from an original image serving as a decoding target image and being a lower dimensional signal than the original image, a first predictive image having a high correlation with the original image, and a second predictive image having a high correlation with the original image with each other to obtain a combined predictive image, and includes an interpolation factor acquisition unit configured to obtain an interpolation factor set for each pixel of the combined predictive image from the first predictive image, the second predictive image, and the observation signal, and a combined predictive image acquisition unit configured to combine the first predictive image and the second predictive image for each pixel using the interpolation factor to obtain the combined predictive image, and the pixels constituting the observation signal are expressed by a
    Type: Grant
    Filed: November 7, 2019
    Date of Patent: December 6, 2022
    Assignee: NIPPON TELEGRAPH AND TELEPHONE CORPORATION
    Inventors: Yoko Sogabe, Shiori Sugimoto, Seishi Takamura, Atsushi Shimizu
  • Patent number: 11516504
    Abstract: Aspects of the disclosure provide methods and apparatuses for video coding. In some examples, an apparatus includes processing circuitry that stores reconstructed samples of a reconstructed block in a memory. When a current sub-block in a current block is to be reconstructed using intra block copy (IBC) based on a reference sub-block in the reconstructed block, the processing circuitry determines whether the reconstructed samples of the reference sub-block stored in the memory are indicated as overwritten based on a position of the current sub-block, generates reconstructed samples of the current sub-block based on the reconstructed samples of the reference sub-block when the reconstructed samples of the reference sub-block stored in the memory are determined to be indicated as not overwritten, and overwrites the reconstructed samples of a collocated sub-block in the reconstructed block stored in the memory with the generated reconstructed samples of the current sub-block.
    Type: Grant
    Filed: October 21, 2020
    Date of Patent: November 29, 2022
    Assignee: TENCENT AMERICA LLC
    Inventors: Xiaozhong Xu, Xiang Li, Shan Liu
  • Patent number: 11509898
    Abstract: There is provided a terminal device capable of efficiently performing communication in a communication system in which a base station device and the terminal device communicate with each other. The terminal device that communicates with the base station device by using a plurality of aggregated cells recognizes that a serving cell is stopped in a first state, recognizes that the serving cell is started in a second state, and switches from the first state to the second state based on a received PDCCH.
    Type: Grant
    Filed: October 22, 2020
    Date of Patent: November 22, 2022
    Assignee: Velos Media, LLC
    Inventors: Tomohiro Ikai, Takeshi Tsukuba
  • Patent number: 11503247
    Abstract: A method of motion estimation and motion compensation for a video processor includes steps of: detecting an input frame rate of a series of input frames; calculating a frame counting value, wherein the frame counting value represents the number of output frame periods between a current input frame and a previous input frame among the series of input frames; calculating a phase step, which is configured to generate a phase coefficient for generating an interpolated frame as an output frame of each of the output frame periods, according to the frame counting value; and generating the interpolated frame based on the current input frame and the previous input frame by using the phase coefficient. Wherein, the step of calculating the frame counting value, the step of calculating the phase step and the step of generating the interpolated frame are consistently performed until the input frame rate is successfully detected.
    Type: Grant
    Filed: January 4, 2022
    Date of Patent: November 15, 2022
    Assignee: NOVATEK Microelectronics Corp.
    Inventors: Chih Chang, I-Feng Lin, Hsiao-En Chang
  • Patent number: 11477476
    Abstract: An example method includes obtaining values of luma motion vectors for a plurality of luma sub-blocks of a current block of the video data selected for coding using affine motion compensation; determining, based on values of luma motion vectors of a sub-set of the plurality of luma sub-blocks, a value of a chroma motion vector for a chroma sub-block that corresponds to the plurality of luma sub-blocks; predicting, using affine motion compensation, respective samples of each luma sub-block of the plurality of luma sub-blocks based on respective values of the luma motion vectors; and predicting, using affine motion compensation, sample the chroma sub-block based on the value of the chroma motion vector.
    Type: Grant
    Filed: October 2, 2019
    Date of Patent: October 18, 2022
    Assignee: QUALCOMM Incorporated
    Inventors: Wei-Jung Chien, Han Huang, Marta Karczewicz, Luong Pham Van, Vadim Seregin
  • Patent number: 11470339
    Abstract: In an example, a method of decoding video data includes generating a residual block of a picture based on a predicted residual block including reconstructing one or more residual values of the residual block based on one or more predicted residual values of the residual block. The method also includes generating a current block of the picture based on a combination of the residual block and a prediction block of the picture.
    Type: Grant
    Filed: August 26, 2014
    Date of Patent: October 11, 2022
    Assignee: QUALCOMM Incorporated
    Inventors: Liwei Guo, Chao Pang, Rajan Laxman Joshi, Joel Sole Rojals, Marta Karczewicz
  • Patent number: 11451767
    Abstract: Provided are method and apparatus methods and apparatuses for encoding an intra-prediction mode based on a secondary IPM. The present invention may comprises a first decision step of determining whether an intra-prediction mode of a current block is included in a first candidate mode set including M candidate modes (M is an integer equal to or greater than 1); a second decision step of determining whether the intra-prediction mode of the current block is included in a second candidate mode set including N candidate modes (N is an integer equal to or greater than 1) based on a first determination result representing a determination result of the first decision step; and an intra prediction mode encoding step of encoding the intra-prediction mode of the current block based on the first determination result or a second determination result representing a determination result of the second decision step.
    Type: Grant
    Filed: November 18, 2016
    Date of Patent: September 20, 2022
    Assignee: Electronics and Telecommunications Research Institute
    Inventors: Dong San Jun, Jung Won Kang, Sung Chang Lim, Hyun Suk Ko, Jin Ho Lee, Hui Yong Kim, Jin Soo Choi
  • Patent number: 11438608
    Abstract: A method of encoding a video data signal (15) is provided, together with a method for decoding. The encoding comprises providing color information (51) for pixels in an image, providing a depth map with depth information (52) for the pixels, providing transition information (56, 57, 60, 70, 71) being representative of a width (63, 73) of a transition region (61, 72) in the image, the transition region (61, 72) comprising a depth transition (62) and blended pixels in which colors of a foreground object and a background object are blended, and generating (24) the video data signal (15) comprising encoded data representing the color information (51), the depth map (52) and the transition information (56, 57, 60, 70, 71). The decoding comprises using the transition information (56, 57, 60, 70, 71) for determining the width (63, 73) of the transition regions (61, 72) and for determining alpha values (53) for pixels inside the transition regions (61, 72).
    Type: Grant
    Filed: September 12, 2018
    Date of Patent: September 6, 2022
    Assignee: Koninklijke Philips N.V.
    Inventors: Wilhelmus Hendrikus Alfonsus Bruls, Christiaan Varekamp, Reinier Bernardus Maria Klein Gunnewiek
  • Patent number: 11426656
    Abstract: A computer-implemented method for online game streaming is provided, which includes receiving a request from a remote client to access a user account. The user account identifies information regarding the user and digital content available for the user to access from the remote client. The request is received over an Internet connection at a hosting service that includes one or more servers and the digital content including a video game. The method includes receiving selection of the game by the hosting service from the remote client and initiating a data rate test of the remote client. The data rate test identifies a maximum available data rate between the remote client and one or more servers of the hosting service. The data rate test includes gradually increasingly the data rate and then gradually decreasing the data rate to identify an active data rate for streaming interactive content of the video game.
    Type: Grant
    Filed: July 21, 2015
    Date of Patent: August 30, 2022
    Assignee: Sony Interactive Entertainment LLC
    Inventors: Roger van der Laan, Stephen G. Perlman
  • Patent number: 11403782
    Abstract: Methods and systems are provided for implementing static channel filtering operations upon image datasets transformed to frequency domain representations, including decoding images of an image dataset to generate a frequency domain representation of the image dataset; discarding coefficient values of one or more particular frequency channels of each image of the image dataset in a frequency domain representation; and transporting the image dataset in a frequency domain representation to one or more special-purpose processor(s). Methods and systems of the present disclosure may enable a filtered image dataset to be input to a second layer of a learning model, bypassing a first layer, or may enable a learning model to be designed with a reduced-size first layer. This may achieve benefits such as reducing computational overhead and time of machine learning training and inference computations, reducing volume of image data input into the learning model, and reducing convergence time.
    Type: Grant
    Filed: December 31, 2019
    Date of Patent: August 2, 2022
    Assignee: Alibaba Group Holding Limited
    Inventors: Kai Xu, Fei Sun, Minghai Qin, Yen-kuang Chen
  • Patent number: 11373407
    Abstract: A computer-implemented method for attention generation is provided. In this method, a plurality of image frames can be obtained from a video stream. An original attention for a first image frame of the plurality of image frames can be generated. Then, at least one interested area can be identified in the first image frame. A local attention for each of the at least one interested area can be generated. Moreover, a total attention for the first image frame can be generated based on the original attention of the first image frame and the local attention of each of the at least one interested area.
    Type: Grant
    Filed: October 25, 2019
    Date of Patent: June 28, 2022
    Assignee: International Business Machines Corporation
    Inventors: Qing Wang, Shi Lei Zhang, Yonghua Lin
  • Patent number: 11368688
    Abstract: An encoder system may include an analyzer that analyzes a current image area in an input video to select a transform. A selectable residue transformer, controlled by the analyzer, may perform the selectable transform on a residue image generated from the current image area and a predicted current image area, to generate a transformed residue image. An encoder may encode the transformed residue image to generate output data. The analyzer controls the encoder to encode information to identify the selectable transform and to indicate that the selectable transform for the current image area is different from a transform of a previous image area of the input video. A decoder system may include components appropriate for decoding the output data from the encoder system.
    Type: Grant
    Filed: March 3, 2020
    Date of Patent: June 21, 2022
    Assignee: Apple Inc.
    Inventor: Alexandros Tourapis
  • Patent number: 11363302
    Abstract: Method and apparatus of video encoding video coding for a video encoder or decoder using Neural Network (NN) are disclosed. According to one method, input data or a video bitstream are received for blocks in one or more pictures, which comprise one or more colour components. The residual data, prediction data, reconstructed data, filtered-reconstructed data or a combination thereof is derived for one or more blocks of said one or more pictures. A target signal corresponding to one or more of the about signal types is processed using a NN (Neural Network) and the input of the NN or an output of the NN comprises two or more colour components. According to another method, A target signal corresponding to one or more of the about signal types is processed using a NN and the input of the NN or an output of the NN comprises two or more colour components.
    Type: Grant
    Filed: September 28, 2018
    Date of Patent: June 14, 2022
    Assignee: MEDIATEK INC.
    Inventors: Yu-Chi Su, Jan Klopp, Ching-Yeh Chen, Tzu-Der Chuang, Yu-Wen Huang
  • Patent number: 11350129
    Abstract: A method for decoding video includes creating a first list of motion vectors from at least one neighboring block in a current frame of the video and creating a second list of motion vectors from at least one previous block in a temporally earlier frame of the video. A third list of motion vectors is created based upon the first list and the second list. Based upon receiving a motion vector competition control parameter selecting one of the motion vectors from the third list, wherein the second list of motion vectors is further based upon a flooring function.
    Type: Grant
    Filed: September 7, 2020
    Date of Patent: May 31, 2022
    Assignee: Velos Media, LLC
    Inventor: Christopher A. Segall
  • Patent number: 11330286
    Abstract: A motion estimation technique finds first and second candidate bi-directional motion vectors for a first region of an interpolated frame of video content by performing double ended vector motion estimation on the first region. One of these candidate bi-directional motion vectors is selected, and used to identify a remote region of the interpolated frame. This remote region is located at an off-set location from the first region, and is found based on an endpoint of the selected candidate bi-directional motion vector. A remote motion vector for the remote region of the interpolated frame is obtained, and one or more properties of this remote motion vector are used to bias a selection between the first and second candidate vectors.
    Type: Grant
    Filed: May 26, 2020
    Date of Patent: May 10, 2022
    Assignee: Imagination Technologies Limited
    Inventor: Steven Fishwick
  • Patent number: 11317098
    Abstract: Apparatus and method for detecting scene changes using data sets gathered from a 3D pipeline. For example, one embodiment of an apparatus comprises: a 3D graphics engine to render a plurality of frames including a current frame and a prior frame; and a scene change detector to retrieve data sets from the 3D graphics engine, the data sets associated with the current frame rendered by the 3D graphics engine, the scene change detector to analyze the data sets in view of corresponding data sets from the prior frame to determine a first type of encoding to be used for encoding the current frame prior to transmission over a network.
    Type: Grant
    Filed: December 7, 2018
    Date of Patent: April 26, 2022
    Assignee: INTEL CORPORATION
    Inventors: Tomasz Madajczak, Jaroslaw Motowidlo, James Varga
  • Patent number: 11281714
    Abstract: An image retrieval method is provided. According to an example of the method, based on a first image feature of an image to be retrieved, a binary feature of the image to be retrieved may be determined, and an information entropy code of the image to be retrieved may be further determined based on the binary feature of the image to be retrieved. Then, based on the information entropy code of the image to be retrieved, images similar to the image to be retrieved may be retrieved in a preset image library.
    Type: Grant
    Filed: October 18, 2018
    Date of Patent: March 22, 2022
    Assignee: Beijing Sankuai Online Technology Co., Ltd
    Inventor: Liping Kang
  • Patent number: 11277638
    Abstract: An image decoding method according to the present document comprises the steps of: deriving transform coefficients through inverse quantization on the basis of quantized transform coefficients for a target block; deriving modified transform coefficients on the basis of an inverse reduced secondary transform (RST) for the transform coefficients; and generating, on the basis of an inverse primary transform for the modified transform coefficients, a restoration picture based on residual samples for the target block, wherein the modified transform coefficients derived according to the inverse RST are two-dimensionally arranged according to the order of a row priority direction or a column priority direction according to an intra prediction mode to be applied to the target block.
    Type: Grant
    Filed: May 25, 2021
    Date of Patent: March 15, 2022
    Assignee: LG ELECTRONICS INC.
    Inventors: Moonmo Koo, Seunghwan Kim, Jaehyun Lim
  • Patent number: 11276186
    Abstract: A captured image is registered, as a background image, from among captured images sequentially inputted. In a case where it is determined that a moving object is included in a first background image registered at a first timing, the registered first background image is replaced with a second background image registered at a second timing prior to the first timing.
    Type: Grant
    Filed: June 13, 2019
    Date of Patent: March 15, 2022
    Assignee: CANON KABUSHIKI KAISHA
    Inventor: Masayuki Kobayashi
  • Patent number: 11277634
    Abstract: The present invention relates to a method and apparatus for processing a video signal, which can increase the accuracy of the motion vector prediction through motion vector scaling which takes a difference in the temporal distance between reference pictures into consideration. To this end, the present invention provides a video signal processing method and a video signal processing apparatus using the same, and the method comprises the steps of: scaling at least one neighboring partition motion vector for a motion vector prediction of the current partition; scaling the neighboring partition motion vector, which has been selected, when the reference picture of the neighboring partition motion vector is different from the reference picture of the current partition; acquiring a motion vector prediction value of the current partition using the scaled motion vector; and acquiring a motion vector of the current partition using the motion vector prediction value.
    Type: Grant
    Filed: July 7, 2020
    Date of Patent: March 15, 2022
    Assignee: LG ELECTRONICS INC.
    Inventors: Joonyoung Park, Seungwook Park, Jaehyun Lim, Jungsun Kim, Younghee Choi, Jaewon Sung, Byeongmoon Jeon, Yongjoon Jeon
  • Patent number: 11272210
    Abstract: Methods of encoding and decoding for video data are describe in which significance maps are encoded and decoded using non-spatially-uniform partitioning of the map into parts, wherein the bit positions within each part are associated with a given context. Example partition sets and processes for selecting from amongst predetermined partition sets and communicating the selection to the decoder are described.
    Type: Grant
    Filed: November 23, 2020
    Date of Patent: March 8, 2022
    Assignee: Velos Media, LLC
    Inventors: Gergely Ferenc Korodi, Jinwen Zan, Dake He
  • Patent number: 11221990
    Abstract: A system for machine learning model parameters for image compression, including partitioning image files into a first set of regions, determining a first set of machine learned model parameters based on the regions, the first set of machine learned model parameters representing a first level of patterns in the image files, constructing a representation of each of the regions based on the first set of machine learned model parameters, constructing representations of the image files by combining the representations of the regions in the first set of regions, partitioning the representations of the image files into a second set of regions, and determining a second set of machine learned model parameters based on the second set of regions, the second set of machine learned model parameters representing a second level of patterns in the image files.
    Type: Grant
    Filed: April 3, 2015
    Date of Patent: January 11, 2022
    Assignee: The MITRE Corporation
    Inventors: Robert M. Taylor, Jr., Burhan Necioglu
  • Patent number: 11223834
    Abstract: Video encoding, and specifically to a method and an encoder which can adapt the probability of intra refresh of image frame subdivisions based on a count of image frames since a previous refresh and modulated through convex or concave shaped distribution functions are set forth herein.
    Type: Grant
    Filed: October 14, 2020
    Date of Patent: January 11, 2022
    Assignee: AXIS AB
    Inventors: Xing Danielsson Fan, Maya Shah, Mattias Pettersson, Adam Schill Collberg, Fredrik Pihl, Magnus Söderdahl, Peter Wilhelmsson
  • Patent number: 11205077
    Abstract: A method is described for operating on a frame of a video to generate a feature map of a neural network. The method determines if a block of the frame is an inter block or an intra block, and performs an inter block process in the event that the block is an inter block and/or an intra block process in the event that the block is an intra block. The inter block process determines a measure of differences between the block of the frame and a reference block of a reference frame of the video, and performs either a first process or a second process based on the measure to generate a segment of the feature map. The intra block process determines a measure of flatness of the block of the frame, and performs either a third process or a fourth process based on the measure to generate a segment of the feature map.
    Type: Grant
    Filed: May 29, 2020
    Date of Patent: December 21, 2021
    Assignee: Arm Limited
    Inventors: Jayavarapu Srinivasa Rao, Daren Croxford, Dominic Hugo Symes
  • Patent number: 11200416
    Abstract: Image detection methods and apparatuses, electronic devices and storage media include: down-sampling a second image to obtain a first image; determining at least one key sub-region in the second image for category detection; respectively extracting a feature of the first image and a feature of the at least one key sub-region; and generating a category detection result of the second image at least based on the extracted features.
    Type: Grant
    Filed: June 28, 2019
    Date of Patent: December 14, 2021
    Assignee: BEIJING SENSETIME TECHNOLOGY DEVELOPMENT CO., LTD
    Inventors: Zhe Wang, Jianping Shi, Hongsheng Li, Xiaogang Wang
  • Patent number: 11200422
    Abstract: A method and a system for pixel channel imbalance compensation for an image sensing circuit are provided. The system includes an image acquisition circuit having a lens, a color filter and an image sensor and a processing circuit. In the method performed by the processing circuit, a second frame image is retrieved from a motion image, and a first frame image that has undergone noise reduction can be retrieved from a memory. Motion detection is performed between the frames by comparing the first frame image and the second frame image. The motion detection is referred to as a reference for determining how to perform 3D noise reduction. A compensation value for channel imbalance between the adjacent channels can be estimated based on the image under noise reduction in a same buffer. While the pixel channel imbalance is compensated, the image is then restored by an interpolation method.
    Type: Grant
    Filed: July 15, 2020
    Date of Patent: December 14, 2021
    Assignee: REALTEK SEMICONDUCTOR CORP.
    Inventors: Wan-Ju Tang, Tsung-Hsuan Li, Shih-Tse Chen
  • Patent number: 11191422
    Abstract: A medical imaging device configured to be inserted to a patient's body, comprising an elongated tube, comprising two or more cameras for capturing visual content in the vicinity of the device in different directions of view, a light source for illuminating a field of view of the two or more cameras, a wireless module configured to transmit data captured by the two or more cameras to a remote devices, and a power source for supplying power to the camera, the light source and the wireless module.
    Type: Grant
    Filed: August 21, 2017
    Date of Patent: December 7, 2021
    Assignee: 270 SURGICAL LTD.
    Inventor: Avraham Levy
  • Patent number: 11184608
    Abstract: The present disclosure relates to an image processing apparatus and a method designed to maintain subjective image quality. An intra prediction unit and a motion compensation unit calculate the costs of the modes of the respective blocks constituting the current frame. An intra refresh determination unit selects modes of the respective blocks, on the basis of the calculated costs and refreshed blocks that are the blocks refreshed in the previous frame that is located before the current frame timewise. The present disclosure can be applied to an image processing apparatus, an image encoding device, an image decoding device, or the like, for example.
    Type: Grant
    Filed: December 4, 2018
    Date of Patent: November 23, 2021
    Assignee: SONY CORPORATION
    Inventor: Hiroyuki Yasuda
  • Patent number: 11182910
    Abstract: In order to detect gross subject movement in a video image in a way which is not sensitive to illumination change, for example illumination changes caused by movement of shadows or sunlight, spaced pairs of image frames are selected from a video sequence and sub-divided into cells, and spatial frequency analysis is performed in each cell. The magnitude of the spatial frequency components in corresponding cells in the two selected image frames are compared. If the number of cells with high magnitude difference is high then the video image is determined as containing gross subject movement whereas if the number of cells with high magnitude differences is low, the sequence is determined as not containing gross movement, though it may contain illumination changes or no or fine movement.
    Type: Grant
    Filed: September 19, 2017
    Date of Patent: November 23, 2021
    Assignee: OXEHEALTH LIMITED
    Inventors: Mohamed Elmikaty, Simon Mark Chave Jones
  • Patent number: 11138697
    Abstract: An image processor (13) of an X-ray imaging apparatus (100) is configured to be switchable to a full period image composition mode in which a composite image (M) is generated by superimposing images in an entirety of an image generation period and a partial period image composition mode in which the composite image is generated by superimposing the images in a part of the image generation period.
    Type: Grant
    Filed: April 9, 2018
    Date of Patent: October 5, 2021
    Assignee: Shimadzu Corporation
    Inventors: Michel Dargis, Takanori Yoshida, Keiichi Tanno
  • Patent number: 11132542
    Abstract: Aspects of the present disclosure describe systems, methods and structures employing time-space de-noising for distributed sensor.
    Type: Grant
    Filed: June 27, 2019
    Date of Patent: September 28, 2021
    Inventors: Philip Nan Ji, Shuji Murakami, Giovanni Milione, Ting Wang
  • Patent number: 11134276
    Abstract: A quantized image frame and a quantization of a high frequency image frame is obtained. The image frame is de-quantized and transformed to produce a low frequency image frame of a first color. A first set of components of the first color and a second set of components of the first color are determined based at least in part on the low frequency image frame and the high frequency image frame. The first set of components of the first color and the second set of components of the first color are combined with a set of components of a second color to produce a set of image data for an image frame. The set of image data is processed to produce the image frame in a particular format to be displayed on a display device.
    Type: Grant
    Filed: January 30, 2019
    Date of Patent: September 28, 2021
    Assignee: Vulcan Inc.
    Inventor: Michael Persiantsev
  • Patent number: 11112796
    Abstract: Systems and methods for predicting object motion and controlling autonomous vehicles are provided. In one example embodiment, a computer implemented method includes obtaining state data indicative of at least a current or a past state of an object that is within a surrounding environment of an autonomous vehicle. The method includes obtaining data associated with a geographic area in which the object is located. The method includes generating a combined data set associated with the object based at least in part on a fusion of the state data and the data associated with the geographic area in which the object is located. The method includes obtaining data indicative of a machine-learned model. The method includes inputting the combined data set into the machine-learned model. The method includes receiving an output from the machine-learned model. The output can be indicative of a plurality of predicted trajectories of the object.
    Type: Grant
    Filed: September 5, 2018
    Date of Patent: September 7, 2021
    Assignee: UATC, LLC
    Inventors: Nemanja Djuric, Vladan Radosavljevic, Thi Duong Nguyen, Tsung-Han Lin, Jeff Schneider, Henggang Cui, Fang-Chieh Chou, Tzu-Kuo Huang
  • Patent number: 11102501
    Abstract: A motion vector field coding and decoding method, where the method includes obtaining an original signal of a current motion vector field block, where the current motion vector field block is obtained by dividing a current motion vector field into blocks, and the current motion vector field is a motion vector field corresponding to a video frame at a moment t, obtaining a prediction signal of the current motion vector field block and prediction information of the current motion vector field block, calculating a prediction residual signal of the current motion vector field block according to the prediction signal and the original signal, where the prediction residual signal is used to indicate a residual between the original signal and the prediction signal, and writing the prediction information and the prediction residual signal into a bitstream.
    Type: Grant
    Filed: February 21, 2018
    Date of Patent: August 24, 2021
    Assignee: HUAWEI TECHNOLOGIES CO., LTD.
    Inventors: Hong Zhang, Haitao Yang
  • Patent number: 11095895
    Abstract: Techniques related to transform coefficient shaping for video encoding are discussed. Such techniques include applying weighting parameters from one or more perceptually-designed matrices of weighting parameters to blocks of transform coefficients to generate weighted transform coefficients and encoding the weighted transform coefficients into a bitstream. The process may be based on sets of perceptually designed matrices of weighting parameters. Classifier outputs may be used to select from the set of perceptually designed matrices a subset of matrices to work with. The latter may be used in a synthesis procedure to develop the final weighting matrix to be used is shaping the transform coefficients.
    Type: Grant
    Filed: February 1, 2018
    Date of Patent: August 17, 2021
    Assignee: Intel Corporation
    Inventors: Nader Mahdi, Chekib Nouira, Hassen Guermazi, Amir Naghdinezhad, Faouzi Kossentini, Foued Ben Amara
  • Patent number: 11082714
    Abstract: The present invention relates to the construction of a search space for determining a motion vector for a current block of a picture in a video sequence. The search space construction is split into two main stages, wherein a first and a second partial search space are respectively determined. Based on an initial estimate of a motion vector, a first search space is first constructed. A first and a second candidate motion vector of the first search space are identified according to a cost function. Based on the first and the second candidate motion vectors, a second search space is constructed. The motion vector for the current block is selected from the candidate motion vectors of the first search space and the second search space.
    Type: Grant
    Filed: December 30, 2019
    Date of Patent: August 3, 2021
    Assignee: Huawei Technologies Co., Ltd.
    Inventors: Semih Esenlik, Anand Meher Kotra, Zhijie Zhao, Han Gao
  • Patent number: 11082713
    Abstract: A method and apparatus of Inter prediction for video coding performed in a video encoder or a video decoder are disclosed. In one method, a GMC (global motion compensation) indication indicating whether global motion compensation is enabled is signalled at the video encoder side or parsed from the video bitstream at the video decoder side. The current processing unit is encoded or decoded using one or more coding modes including a GMC mode if the GMC indication is indicates GMC is enabled. According to another method, if a GMC mode is enabled for a region, a number representing the total number of multiple global motion models allowed is determined for encoding or decoding the processing units in the region using multiple global motion models. In yet another method the GMC and local motion compensation are applied selectively or in a combined fashion.
    Type: Grant
    Filed: November 18, 2016
    Date of Patent: August 3, 2021
    Assignee: MEDIATEK INC.
    Inventors: Shan Liu, Xiaozhong Xu
  • Patent number: 11070839
    Abstract: A further coding efficiency increase is achieved by, in hybrid video coding, additionally predicting the residual signal of a current frame by motion-compensated prediction using a reference residual signal of a previous frame. In other words, in order to further reduce the energy of the final residual signal, i.e. the one finally transmitted, and thus increase the coding efficiency, it is proposed to additionally predict the residual signal by motion-compensated prediction using the reconstructed residual signals of previously coded frames.
    Type: Grant
    Filed: May 4, 2012
    Date of Patent: July 20, 2021
    Assignee: FRAUNHOFER-GESELLSCHAFT ZUR FOERDERUNG DER ANGEWANDTEN FORSCHUNG E.V.
    Inventors: Benjamin Bross, Thomas Wiegand, Heiko Schwarz
  • Patent number: 11055869
    Abstract: The present disclosure discloses a method, apparatus and system for assisting security inspection, and relates to the field of security inspection. The method includes: acquiring registration information of an inspected object; acquiring a standard scanned image corresponding to the registration information; displaying the standard scanned image in an AR manner to determine whether the inspected object is a suspicious object through comparing the standard scanned image with an actual scanned image, the actual scanned image comprising an image of the inspected object.
    Type: Grant
    Filed: December 20, 2018
    Date of Patent: July 6, 2021
    Inventors: Bicheng Liu, Haoran Zhang, Guangming Xu, Qi Wang, Qiangqiang Zhu, Yuan Wo