公开(公告)号：US12061991B2

公开(公告)日：2024-08-13

申请号：US17029506

申请日：2020-09-23

申请人： International Business Machines Corporation ,  National Tsing Hua University

发明人： Pin-Yu Chen ,  Sijia Liu ,  Chia-Yu Chen ,  I-Hsin Chung ,  Tsung-Yi Ho ,  Yun-Yun Tsai

IPC分类号： G06N3/094 ,  G06N3/08 ,  G06N3/096 ,  G06N20/00 ,  G06F18/213 ,  G06F18/2134 ,  G06F18/214

CPC分类号： G06N3/094 ,  G06N3/08 ,  G06N3/096 ,  G06N20/00 ,  G06F18/213 ,  G06F18/21347 ,  G06F18/214

摘要： Transfer learning in machine learning can include receiving a machine learning model. Target domain training data for reprogramming the machine learning model using transfer learning can be received. The target domain training data can be transformed by performing a transformation function on the target domain training data. Output labels of the machine learning model can be mapped to target labels associated with the target domain training data. The transformation function can be trained by optimizing a parameter of the transformation function. The machine learning model can be reprogrammed based on input data transformed by the transformation function and a mapping of the output labels to target labels.

公开(公告)号：US11681000B2

公开(公告)日：2023-06-20

申请号：US17478127

申请日：2021-09-17

申请人： Hyperfine Operations, Inc.

发明人： Jo Schlemper ,  Seyed Sadegh Mohseni Salehi ,  Michal Sofka

IPC分类号： G01R33/56 ,  G01R33/561 ,  A61B5/055 ,  G06N3/082 ,  G06T7/38 ,  G01R33/383 ,  G01R33/44 ,  G06N3/08 ,  G06T3/60 ,  G06T11/00 ,  G16H30/40 ,  G01R33/36 ,  G06T7/262 ,  G06T7/00 ,  G06V10/75 ,  G06F18/2134 ,  G06N3/045 ,  G06V10/42 ,  G06V10/82 ,  G06V10/44 ,  G06V10/30 ,  G06V10/52

CPC分类号： G01R33/5611 ,  A61B5/055 ,  G01R33/36 ,  G01R33/383 ,  G01R33/445 ,  G01R33/5608 ,  G06F18/21347 ,  G06N3/045 ,  G06N3/08 ,  G06N3/082 ,  G06T3/60 ,  G06T7/0012 ,  G06T7/262 ,  G06T7/38 ,  G06T11/006 ,  G06T11/008 ,  G06V10/30 ,  G06V10/431 ,  G06V10/454 ,  G06V10/52 ,  G06V10/7515 ,  G06V10/82 ,  G16H30/40 ,  G06T2207/10088 ,  G06T2207/20056 ,  G06T2207/20081 ,  G06T2207/20084 ,  G06T2207/20182 ,  G06T2207/20216 ,  G06T2207/20224 ,  G06T2207/30016 ,  G06T2210/41

摘要： Techniques for generating magnetic resonance (MR) images of a subject from MR data obtained by a magnetic resonance imaging (MRI) system, the techniques including: obtaining input MR data obtained by imaging the subject using the MRI system; generating a plurality of transformed input MR data instances by applying a respective first plurality of transformations to the input MR data; generating a plurality of MR images from the plurality of transformed input MR data instances and the input MR data using a non-linear MR image reconstruction technique; generating an ensembled MR image from the plurality of MR images at least in part by: applying a second plurality of transformations to the plurality of MR images to obtain a plurality of transformed MR images; and combining the plurality of transformed MR images to obtain the ensembled MR image; and outputting the ensembled MR image.

公开(公告)号：US12105173B2

公开(公告)日：2024-10-01

申请号：US18312654

申请日：2023-05-05

申请人： Hyperfine Operations, Inc.

发明人： Jo Schlemper ,  Seyed Sadegh Mohseni Salehi ,  Michal Sofka

IPC分类号： G01R33/561 ,  A61B5/055 ,  G01R33/36 ,  G01R33/383 ,  G01R33/44 ,  G01R33/56 ,  G06F18/2134 ,  G06N3/045 ,  G06N3/08 ,  G06N3/082 ,  G06T3/60 ,  G06T7/00 ,  G06T7/262 ,  G06T7/38 ,  G06T11/00 ,  G06V10/30 ,  G06V10/42 ,  G06V10/44 ,  G06V10/52 ,  G06V10/75 ,  G06V10/82 ,  G06V10/88 ,  G16H30/40

CPC分类号： G01R33/5611 ,  A61B5/055 ,  G01R33/36 ,  G01R33/383 ,  G01R33/445 ,  G01R33/5608 ,  G06F18/21347 ,  G06N3/045 ,  G06N3/08 ,  G06N3/082 ,  G06T3/60 ,  G06T7/0012 ,  G06T7/262 ,  G06T7/38 ,  G06T11/006 ,  G06T11/008 ,  G06V10/30 ,  G06V10/431 ,  G06V10/454 ,  G06V10/52 ,  G06V10/7515 ,  G06V10/82 ,  G06V10/89 ,  G06V10/92 ,  G16H30/40 ,  G06T2207/10088 ,  G06T2207/20056 ,  G06T2207/20081 ,  G06T2207/20084 ,  G06T2207/20182 ,  G06T2207/20216 ,  G06T2207/20224 ,  G06T2207/30016 ,  G06T2210/41

摘要： Techniques for generating magnetic resonance (MR) images of a subject from MR data obtained by a magnetic resonance imaging (MRI) system, the techniques including: obtaining input MR data obtained by imaging the subject using the MRI system; generating a plurality of transformed input MR data instances by applying a respective first plurality of transformations to the input MR data; generating a plurality of MR images from the plurality of transformed input MR data instances and the input MR data using a non-linear MR image reconstruction technique; generating an ensembled MR image from the plurality of MR images at least in part by: applying a second plurality of transformations to the plurality of MR images to obtain a plurality of transformed MR images; and combining the plurality of transformed MR images to obtain the ensembled MR image; and outputting the ensembled MR image.

公开(公告)号：US11907850B2

公开(公告)日：2024-02-20

申请号：US17454516

申请日：2021-11-11

申请人： Google LLC

发明人： Rui Zhang ,  Jia Li ,  Tomas Jon Pfister

IPC分类号： G06N3/084 ,  G06F18/214 ,  G06F18/22 ,  G06F18/21 ,  G06F18/2413 ,  G06F18/2134 ,  G06N3/045 ,  G06N3/047 ,  G06V10/74 ,  G06V10/764 ,  G06V10/774 ,  G06V10/82

CPC分类号： G06N3/084 ,  G06F18/214 ,  G06F18/2148 ,  G06F18/2193 ,  G06F18/21347 ,  G06F18/22 ,  G06F18/2413 ,  G06N3/045 ,  G06N3/047 ,  G06V10/761 ,  G06V10/764 ,  G06V10/774 ,  G06V10/82

摘要： A method includes obtaining a source training dataset that includes a plurality of source training images and obtaining a target training dataset that includes a plurality of target training images. For each source training image, the method includes translating, using the forward generator neural network G, the source training image to a respective translated target image according to current values of forward generator parameters. For each target training image, the method includes translating, using a backward generator neural network F, the target training image to a respective translated source image according to current values of backward generator parameters. The method also includes training the forward generator neural network G jointly with the backward generator neural network F by adjusting the current values of the forward generator parameters and the backward generator parameters to optimize an objective function.

公开(公告)号：US20240160937A1

公开(公告)日：2024-05-16

申请号：US18418197

申请日：2024-01-19

申请人： Google LLC

发明人： Rui Zhang ,  Jia Li ,  Tomas Jon Pfister

IPC分类号： G06N3/084 ,  G06F18/21 ,  G06F18/2134 ,  G06F18/214 ,  G06F18/22 ,  G06F18/2413 ,  G06N3/045 ,  G06N3/047 ,  G06V10/74 ,  G06V10/764 ,  G06V10/774 ,  G06V10/82

CPC分类号： G06N3/084 ,  G06F18/21347 ,  G06F18/214 ,  G06F18/2148 ,  G06F18/2193 ,  G06F18/22 ,  G06F18/2413 ,  G06N3/045 ,  G06N3/047 ,  G06V10/761 ,  G06V10/764 ,  G06V10/774 ,  G06V10/82

摘要： A method includes obtaining a source training dataset that includes a plurality of source training images and obtaining a target training dataset that includes a plurality of target training images. For each source training image, the method includes translating, using the forward generator neural network G, the source training image to a respective translated target image according to current values of forward generator parameters. For each target training image, the method includes translating, using a backward generator neural network F, the target training image to a respective translated source image according to current values of backward generator parameters. The method also includes training the forward generator neural network G jointly with the backward generator neural network F by adjusting the current values of the forward generator parameters and the backward generator parameters to optimize an objective function.

公开(公告)号：US20240118359A1

公开(公告)日：2024-04-11

申请号：US18312654

申请日：2023-05-05

申请人： Hyperfine Operations, Inc.

发明人： Jo Schlemper ,  Seyed Sadegh Mohseni Salehi ,  Michal Sofka

IPC分类号： G01R33/561 ,  A61B5/055 ,  G01R33/36 ,  G01R33/383 ,  G01R33/44 ,  G01R33/56 ,  G06F18/2134 ,  G06N3/045 ,  G06N3/08 ,  G06N3/082 ,  G06T3/60 ,  G06T7/00 ,  G06T7/262 ,  G06T7/38 ,  G06T11/00 ,  G06V10/30 ,  G06V10/42 ,  G06V10/44 ,  G06V10/52 ,  G06V10/75 ,  G06V10/82 ,  G06V10/88 ,  G16H30/40

CPC分类号： G01R33/5611 ,  A61B5/055 ,  G01R33/36 ,  G01R33/383 ,  G01R33/445 ,  G01R33/5608 ,  G06F18/21347 ,  G06N3/045 ,  G06N3/08 ,  G06N3/082 ,  G06T3/60 ,  G06T7/0012 ,  G06T7/262 ,  G06T7/38 ,  G06T11/006 ,  G06T11/008 ,  G06V10/30 ,  G06V10/431 ,  G06V10/454 ,  G06V10/52 ,  G06V10/7515 ,  G06V10/82 ,  G06V10/89 ,  G06V10/92 ,  G16H30/40 ,  G06T2207/10088 ,  G06T2207/20056 ,  G06T2207/20081 ,  G06T2207/20084 ,  G06T2207/20182 ,  G06T2207/20216 ,  G06T2207/20224 ,  G06T2207/30016 ,  G06T2210/41

摘要： Techniques for generating magnetic resonance (MR) images of a subject from MR data obtained by a magnetic resonance imaging (MRI) system, the techniques including: obtaining input MR data obtained by imaging the subject using the MRI system; generating a plurality of transformed input MR data instances by applying a respective first plurality of transformations to the input MR data; generating a plurality of MR images from the plurality of transformed input MR data instances and the input MR data using a non-linear MR image reconstruction technique; generating an ensembled MR image from the plurality of MR images at least in part by: applying a second plurality of transformations to the plurality of MR images to obtain a plurality of transformed MR images; and combining the plurality of transformed MR images to obtain the ensembled MR image; and outputting the ensembled MR image.

公开(公告)号：US11790038B2

公开(公告)日：2023-10-17

申请号：US17528129

申请日：2021-11-16

申请人： Waymo LLC

发明人： Xiaohan Jin ,  Junhua Mao ,  Ruizhongtai Qi ,  Congcong Li ,  Huayi Zeng

IPC分类号： G06F18/2134 ,  G06T15/06 ,  G06T19/20 ,  G06N3/08 ,  G06T17/20 ,  G06V20/56 ,  G06V40/10 ,  G06F18/214 ,  G06N3/088 ,  G06V10/82 ,  G06T17/00 ,  G06N3/045 ,  G06V20/64

CPC分类号： G06F18/21347 ,  G06F18/2148 ,  G06N3/08 ,  G06T15/06 ,  G06T17/20 ,  G06T19/20 ,  G06V20/56 ,  G06V40/103 ,  G06F18/214 ,  G06N3/045 ,  G06N3/088 ,  G06T17/00 ,  G06T2210/56 ,  G06T2219/2004 ,  G06V10/82 ,  G06V20/64

摘要： Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating rare pose data. One of the methods includes obtaining a three-dimensional model of a dynamic object, wherein the dynamic object has multiple movable elements that define a plurality of poses of the dynamic object. A plurality of template poses of the dynamic object are used to generate additional poses for the dynamic object including varying angles of one or more key joints of the dynamic object according to the three-dimensional model. Point cloud data is generated for the additional poses generated for the dynamic object.