Patents Examined by Chan S. Park
-
Patent number: 11810303Abstract: The present application relates to an automated segmentation method for use with echocardiograms (echo). The method uses an iterative Dijkstra's algorithm, a strategic node selection, and a novel cost matrix formulation based on intensity peak prominence and is this termed the “Prominence Iterative Dijkstra's” algorithm, or ProID.Type: GrantFiled: March 11, 2021Date of Patent: November 7, 2023Assignee: Purdue Research FoundationInventors: Brett Albert Meyers, Pavlos P. Vlachos, Melissa Brindise
-
Patent number: 11804059Abstract: The present disclosure provides a pedestrian re-identification method and apparatus. The pedestrian re-identification apparatus detects a pedestrian to be identified in a specified video frame, and within a specified time period before the capture time of the specified video frame, extracts all pedestrians appearing in the captured video frame and taking them to be historical pedestrians; calculating the feature distances between the pedestrian to be identified and the historical pedestrians; in order of feature distance, from small to large, extracting historical pedestrian identifications corresponding to a predetermined number of feature distances, and clustering the extracted historical pedestrian identifications so as to re-identify the pedestrians to be identified. The present disclosure effectively suppresses external noise interference and improves the success rate of pedestrian re-identification.Type: GrantFiled: April 26, 2019Date of Patent: October 31, 2023Assignees: BEIJING JINGDONG SHANGKE INFORMATION TECHNOLOGY CO., LTD., BEIJING JINGDONG CENTURY TRADING CO., LTD.Inventors: Jiangxu Wu, Weihua Zhang, Hongguang Zhang, Ganglin Peng, Leifeng Kong, Yuefeng Wu, Xinghui Yong, Fan Li
-
Patent number: 11803758Abstract: This document relates to training of machine learning models. One example method involves providing a machine learning model having one or more mapping layers. The one or more mapping layers can include at least a first mapping layer configured to map components of pretraining examples into first representations in a space. The example method also includes performing a pretraining stage on the one or more mapping layers using the pretraining examples. The pretraining stage can include adding noise to the first representations of the components of the pretraining examples to obtain noise-adjusted first representations. The pretraining stage can also include performing a self-supervised learning process to pretrain the one or more mapping layers using at least the first representations of the training data items and the noise-adjusted first representations of the training data items.Type: GrantFiled: May 22, 2020Date of Patent: October 31, 2023Assignee: Microsoft Technology Licensing, LLCInventors: Xiaodong Liu, Hao Cheng, Yu Wang, Jianfeng Gao, Weizhu Chen, Pengcheng He, Hoifung Poon
-
Patent number: 11776145Abstract: In one embodiment, a method includes receiving, by an object reconstruction module, a first image and a second image. The first image includes a first region of an object and the second image comprises a second region of the object. The method also includes identifying, by the object reconstruction module, a transitional image. The transitional image includes the first region of the object and the second region of the object. The method further includes determining, by the object reconstruction module, that the first region of the object in the transitional image and the first region of the object in the first image are equivalent regions and generating, by the object reconstruction module, a reconstruction of the object using the first image and the transitional image. The reconstruction of the object includes the first region of the object and the second region of the object and excludes equivalent regions.Type: GrantFiled: July 20, 2022Date of Patent: October 3, 2023Assignee: BNSF Railway CompanyInventor: Rachel Kohler
-
Patent number: 11775835Abstract: Systems and methods for estimating a layout of a room are disclosed. The room layout can comprise the location of a floor, one or more walls, and a ceiling. In one aspect, a neural network can analyze an image of a portion of a room to determine the room layout. The neural network can comprise a convolutional neural network having an encoder sub-network, a decoder sub-network, and a side sub-network. The neural network can determine a three-dimensional room layout using two-dimensional ordered keypoints associated with a room type. The room layout can be used in applications such as augmented or mixed reality, robotics, autonomous indoor navigation, etc.Type: GrantFiled: April 9, 2020Date of Patent: October 3, 2023Assignee: MAGIC LEAP, INC.Inventors: Chen-Yu Lee, Vijay Badrinarayanan, Tomasz Jan Malisiewicz, Andrew Rabinovich
-
Patent number: 11776127Abstract: Described is a method for segmenting measurement data from measurement of an object that has at least one material transition region. The measurement data are used to generate a digital object representation that has the material transition region and a multiplicity of spatially resolved image information items relating to the object. The method may include: determining measurement data that have at least one small structure having an extent which is less than a predefined extent; determining at least two homogeneous regions in the measurement data and/or in the digital object representation, wherein at least one homogeneous regions has a small structure; analysing a local similarity of the multiplicity of spatially resolved image information items; adapting an extent of each homogeneous region until at least one border region is arranged at an expected position of a material transition region; and segmenting the digital object representation based on the adapted homogeneous regions.Type: GrantFiled: November 20, 2020Date of Patent: October 3, 2023Assignee: VOLUME GRAPHICS GMBHInventors: Christoph Poliwoda, Sören Schüller
-
Patent number: 11762394Abstract: A position detection apparatus comprises: a map information creation part that creates map information representing a position in a space including a floor surface on which at least one detection object may be disposed; a mask information creation part that creates mask information, by extracting a region of a predetermined height range within a height range from the floor surface when the detection object is disposed in the space from the map information; a specifying part that specifies partial optical image information, by removing a region corresponding to the mask information from the optical image information; and a detection part that specifies a positional relationship between the map information and the partial optical image information at a pixel level, and detects a position of the detection object in the map information based on the specified positional relationship.Type: GrantFiled: October 30, 2019Date of Patent: September 19, 2023Assignee: NEC CORPORATIONInventors: Shinya Yasuda, Hiroshi Yoshida
-
Patent number: 11763425Abstract: Techniques and systems are provided for high resolution time-of-flight (ToF) depth imaging. In some examples, an apparatus includes a projection system including one or more light-emitting devices, each light-emitting device being configured to illuminate at least one portion of an entire field-of-view (FOV) of the projection system. The entire FOV includes a plurality of FOV portions. The apparatus also includes a receiving system including a sensor configured to sequentially capture a plurality of images based on a plurality of illumination reflections corresponding to light emitted by the one or more light-emitting devices. Each image of the plurality of images corresponds to one of the plurality of FOV portions. An image resolution associated with each image corresponds to a full resolution of the sensor. The apparatus further includes a processor configured to generate, using the plurality of images, an increased resolution depth map associated with the entire FOV.Type: GrantFiled: February 3, 2021Date of Patent: September 19, 2023Assignee: QUALCOMM IncorporatedInventors: Jian Ma, Sergiu Radu Goma, Biay-Cheng Hseih
-
Patent number: 11756216Abstract: A system includes sensors and a processor that tracks first and second objects in a space. Upon detecting that a tracked position of the first object is within a threshold distance of a tracked position of a second object, a top-view image of the first object is received from a first sensor. Based on this top-view image, a first descriptor is determined for the first object. The first descriptor is associated with an observable characteristic of the first object. The processor identifies the first object based at least in part upon the first descriptor.Type: GrantFiled: May 21, 2021Date of Patent: September 12, 2023Assignee: 7-ELEVEN, INC.Inventors: Shahmeer Ali Mirza, Sailesh Bharathwaaj Krishnamurthy, Madan Mohan Chinnam, Crystal Maung
-
Patent number: 11741615Abstract: A map segmentation method and device, a motion estimation method, and a device terminal. Obtaining all map point data and constructing a corresponding graph based on these map points, where the graph includes each node formed by map points and corresponding edges; measuring and estimating to-be-estimated state variables corresponding to the node to obtain a corresponding measured and estimated value, where the to-be-estimated state variables include homogeneous coordinates of the node and a corresponding sensor pose upon node obtaining; calculating an inter-point correlation between any two adjacent nodes in the graph based on the measured and estimated value to determine whether the inter-point correlation is less than a preset correlation threshold; and when the inter-point correlation is less than the preset correlation threshold, removing an edge corresponding to the corresponding adjacent nodes, to separate the static points and the dynamic points and complete map segmentation.Type: GrantFiled: April 2, 2021Date of Patent: August 29, 2023Assignee: ZHEJIANG UNIVERSITYInventors: Yu Zhang, Weichen Dai, Jiajun Jiang
-
Patent number: 11734840Abstract: The present disclosure relates to a target tracking method, device, medium and apparatus, belongs to the technical field of computers, and can ensure the completeness of tracking. The target tracking method comprises performing object detection on all telephoto image frames captured by telephoto cameras in a multi-camera system, and constructing a detection point set based on object detection result for each frame of the telephoto image; selecting a target tracking point from each detection point set based on the starting tracking point; and connecting the selected target tracking points as a tracking sequence for the starting tracking point.Type: GrantFiled: April 20, 2021Date of Patent: August 22, 2023Assignee: BOYAN TECHNOLOGIES (SHENZHEN) CO., LTDInventors: Yuxing Han, Jiangtao Wen, Jisheng Li, Yubin Hu
-
Patent number: 11727588Abstract: A method for depth estimation performed by a depth estimation system of an autonomous agent includes determining a first pose of a sensor based on a first image captured by the sensor and a second image captured by the sensor. The method also includes determining a first depth of the first image and a second depth of the second image. The method further includes generating a warped depth image based on at least the first depth and the first pose. The method still further includes determining a second pose based on the warped depth image and the second depth image. The method also includes updating the first pose based on the second pose and updating a first warped image based on the updated first pose.Type: GrantFiled: April 14, 2021Date of Patent: August 15, 2023Assignee: TOYOTA RESEARCH INSTITUTE, INC.Inventors: Jiexiong Tang, Rares Andrei Ambrus, Vitor Guizilini, Adrien David Gaidon
-
Patent number: 11727589Abstract: A method for multi-camera monocular depth estimation using pose averaging is described. The method includes determining a multi-camera photometric loss associated with a multi-camera rig of an ego vehicle. The method also includes determining a multi-camera pose consistency constraint (PCC) loss associated with the multi-camera rig of the ego vehicle. The method further includes adjusting the multi-camera photometric loss according to the multi-camera PCC loss to form a multi-camera PCC photometric loss. The method also includes training a multi-camera depth estimation model and an ego-motion estimation model according to the multi-camera PCC photometric loss. The method further includes predicting a 360° point cloud of a scene surrounding the ego vehicle according to the trained multi-camera depth estimation model and the ego-motion estimation model.Type: GrantFiled: July 16, 2021Date of Patent: August 15, 2023Assignee: TOYOTA RESEARCH INSTITUTE, INC.Inventors: Vitor Guizilini, Rares Andrei Ambrus, Adrien David Gaidon, Igor Vasiljevic, Gregory Shakhnarovich
-
Patent number: 11720793Abstract: In one aspect, a method includes obtaining videos and for each video: obtaining a set of anchors for the video, each anchor beginning at the playback time and including anchor text; identifying, from text generated from audio of the video, a set of entities specified in the text, wherein each entity in the set of entities is associated with a times stamp at which the entity is mentioned; determining, by a language model and from the text generated from the audio of the video, an importance value for each entity; for a subset of the videos, receiving rater data that describes, for each anchor, the accuracy of the anchor text in describing subject matter of the video; and training, using the human rater data, the importance values, the text, and the set of entities, an anchor model that predicts an entity label for an anchor for a video.Type: GrantFiled: October 13, 2020Date of Patent: August 8, 2023Assignee: GOOGLE LLCInventors: Gabe Culbertson, Wei Peng, Nicolas Crowell
-
Patent number: 11715218Abstract: An information processing obtains a first image which is obtained by using a first camera to capture a three-dimensional space; obtains map information that is generated based on a second transformed image obtained by using second transformation information to geometrically transform a second image which has been captured by using a second camera to capture the three-dimensional space; generates a first transformed image by using first transformation information to geometrically transform the first image; and collates the map information and the first transformed image. The first transformation information and the second transformation information are set so that a geometric correspondence relationship between the three-dimensional space and the first transformed image and a geometric correspondence relationship between the three-dimensional space and the second transformed image will have a common geometric correspondence relationship.Type: GrantFiled: June 28, 2021Date of Patent: August 1, 2023Assignee: Canon Kabushiki KaishaInventor: Makoto Tomioka
-
Patent number: 11669992Abstract: A computer-implemented method comprises obtaining first data comprising a first collection of coordinate data representing a position, or positions, of a first group of one or more objects detected in a first frame of a scene and obtaining second data comprising a second collection of coordinate data representing a position, or positions, of a second group of one or more objects detected in a second frame of the scene, wherein the second frame represents a different view of the scene than the first frame. The method further comprises determining whether any of the first group of objects correspond to any of the second group of objects in the scene based on the first collection of coordinate data and the second collection of coordinate data, wherein the first frame and the second frame are obtained from image data captured from a single camera position.Type: GrantFiled: January 29, 2021Date of Patent: June 6, 2023Assignee: SEECHANGE TECHNOLOGIES LIMITEDInventors: Koki Mitsunami, Michael Andrew Pallister
-
Patent number: 11657477Abstract: An image processing device, an image processing system, an imaging device, an image processing method, and a recording medium storing program code. The image processing device includes a receiver configured to receive a first image and a second image, a memory configured to store a joining position at each one of a plurality of portions between the first image and the second image received by the receiver, and circuitry configured to evaluate, for each of the plurality of portions, validity of the joining position in a past stored in the memory, and generate a composite image based on the first image and the second image received by the receiver and the joining position in the past stored in the memory on a site where the validity meets a criterion. The method includes reading a past joining position at each one of a plurality of portions between a pair of images.Type: GrantFiled: January 26, 2021Date of Patent: May 23, 2023Assignee: RICOH COMPANY, LTD.Inventors: Kazuhiro Yoshida, Takamichi Katoh, Takashi Tada, Hiroyuki Kawamoto, Makoto Usami, Kenichi Ono
-
Patent number: 11645831Abstract: There is provided a method and a system configured obtain an image of a semiconductor specimen including one or more arrays, each including repetitive structural elements, and one or more regions, each region at least partially surrounding a corresponding array and including features different from the repetitive structural elements, wherein the PMC is configured to, during run-time scanning of the semiconductor specimen, perform a correlation analysis between pixel intensity of the image and pixel intensity of a reference image informative of at least one of the repetitive structural elements, to obtain a correlation matrix, use the correlation matrix to distinguish between one or more first areas of the image corresponding to the one or more arrays and one or more second areas of the image corresponding the one or more regions, and output data informative of the one or more first areas of the image.Type: GrantFiled: July 7, 2020Date of Patent: May 9, 2023Assignee: Applied Materials Israel Ltd.Inventors: Yehuda Cohen, Rafael Bistritzer
-
Patent number: 11645769Abstract: A method of determining macrotexture of an object is presented which includes obtaining a plurality of stereo images from an object, generating a local coordinate system for each image, detecting one or more local keypoints each having a local coordinate, generating a global coordinate system based on a plurality of ground control points (GCPs) with apriori position knowledge of each of the plurality of GCPs, transforming the one or more local keypoints in each image to one or more global keypoints each having a global coordinate, generating a sparse point cloud based on the one or more global keypoints, reconstructing a 3D dense point cloud of the object based on neighboring pixels of each of the one or more local keypoints and calculating the global coordinates of each pixel of the 3D dense point cloud, and obtaining the macrotexture based on the reconstructed 3D dense point cloud of the object.Type: GrantFiled: March 15, 2021Date of Patent: May 9, 2023Assignee: Purdue Research FoundationInventors: Jie Shan, Xiangxi Tian
-
Patent number: 11640688Abstract: A data processing apparatus comprises a processing unit to generate image data for rendering data for display as an image frame, a memory to store the image data, the image data comprising data for a given object in the image frame, detection circuitry to detect, for the data for the given object, one or more properties associated with the given object in the image frame, prediction circuitry to generate, based on one or more of the properties, a confidence score for the data for the given object indicative of a likelihood that the data for the given object is to be used for more than a threshold number of image frames in a sequence of image frames, and allocation circuitry to allocate indicator data to the data for the given object, the indicator data indicative of a magnitude of the confidence score for the data for the given object, and to store the data for the given object in the memory in association with the corresponding indicator data.Type: GrantFiled: July 13, 2020Date of Patent: May 2, 2023Assignee: Sony Interactive Entertainment Inc.Inventor: Phillip David Power