公开(公告)号：US20160086357A1

公开(公告)日：2016-03-24

申请号：US14862363

申请日：2015-09-23

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Jun-young PARK ,  Nasir DESAI ,  Yeon-mo JEONG ,  Abhinav MEHROTRA

IPC: G06T11/00 ,  G06T1/20

CPC classification number: G06T11/006 ,  G06T2211/412

Abstract: An apparatus for processing a medical image includes an image processor including a plurality of processors, the plurality of processors configured to reconstruct a cross-sectional image of an object by performing a first operation having a first priority and a second operation having a second priority that is lower than the first priority, and a controller configured to monitor whether a malfunction occurs among the plurality of processors, and configured to assign, to at least one of the plurality of processors, at least one of the first operation and the second operation to be performed, based on a result of monitoring of the plurality of processors.

Abstract translation: 一种用于处理医学图像的装置包括：图像处理器，包括多个处理器，所述多个处理器被配置为通过执行具有第一优先级的第一操作和具有第二优先级的第二操作来重建对象的横截面图像， 低于第一优先级;以及控制器，被配置为监视在所述多个处理器之间是否发生故障，并且被配置为向所述多个处理器中的至少一个处理器分配所述第一操作和所述第二操作中的至少一个， 基于多个处理器的监视结果来执行。

公开(公告)号：US20230410818A1

公开(公告)日：2023-12-21

申请号：US18220567

申请日：2023-07-11

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Alberto Gil C. P. RAMOS ,  Abhinav MEHROTRA ,  Sourav BHATTACHARYA

IPC: G10L17/18 ,  G10L17/04 ,  G10L21/0208 ,  G10L13/08

CPC classification number: G10L17/18 ,  G10L13/08 ,  G10L21/0208 ,  G10L17/04

Abstract: Broadly speaking, embodiments of the present techniques provide a method and system for personalising machine learning models on resource-constrained devices by using conditional neural networks. In particular, the present techniques allow for resource-efficient use of a conditioning vector by incorporating the conditioning vector into weights learned during training. This reduces the computational resources required at inference time.

公开(公告)号：US20220101089A1

公开(公告)日：2022-03-31

申请号：US17477851

申请日：2021-09-17

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Mohamed Saied Abdelkader ABDELFATTAH ,  Abhinav MEHROTRA ,  Lukasz DUDZIAK

IPC: G06N3/04 ,  G06F17/18 ,  G06N3/08

Abstract: The disclosure relates to methods, apparatuses and systems for improving a neural architecture search (NAS). For example, A computer-implemented method using a searching algorithm to design a neural network architecture is provided, the method including: obtaining a plurality of neural network models; selecting a first subset of the plurality of neural network models; applying the searching algorithm to the selected subset of models; and identifying an optimal neural network architecture by repeating the selecting and applying for a fixed number of iterations; wherein at least one score indicative of validation loss for each model is used in or alongside at least one of the selecting and applying.

公开(公告)号：US20180253870A1

公开(公告)日：2018-09-06

申请号：US15971333

申请日：2018-05-04

Applicant: Samsung Electronics Co., Ltd.

Inventor： Jun-young PARK ,  Nasir DESAI ,  Yeon-mo JEONG ,  Abhinav MEHROTRA

IPC: G06T11/00 ,  G06T7/00 ,  G06F11/20

CPC classification number: G06T11/003 ,  G06F11/2028 ,  G06T7/0012 ,  G06T2211/428

Abstract: An apparatus for processing a medical image includes an image processor including a plurality of processors, the plurality of processors configured to reconstruct a cross-sectional image of an object by performing a first operation having a first priority and a second operation having a second priority that is lower than the first priority, and a controller configured to monitor whether a malfunction occurs among the plurality of processors, and configured to assign, to at least one of the plurality of processors, at least one of the first operation and the second operation to be performed, based on a result of monitoring of the plurality of processors.

公开(公告)号：US20220375462A1

公开(公告)日：2022-11-24

申请号：US17874826

申请日：2022-07-27

Applicant: Samsung Electronics Co., Ltd.

Inventor： Sourav BHATTACHARYA ,  Abhinav MEHROTRA ,  Alberto Gil C. P. RAMOS

IPC: G10L15/16 ,  G10L21/0208

Abstract: Broadly speaking, the present techniques provide methods for conditioning a neural network, which not only improve the generalizable performance of conditional neural networks, but also reduce model size and latency significantly. The resulting conditioned neural network is suitable for on-device deployment due to having a significantly lower model size, lower dynamic memory requirement, and lower latency.

公开(公告)号：US20160120497A1

公开(公告)日：2016-05-05

申请号：US14706078

申请日：2015-05-07

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Desai NASIR ,  Abhinav MEHROTRA

IPC: A61B6/00 ,  G06T11/60

CPC classification number: A61B6/56 ,  A61B6/06 ,  A61B6/4291 ,  A61B6/461 ,  A61B6/54 ,  A61B6/586 ,  G06F11/0736 ,  G06F11/0766 ,  G06F11/32 ,  G16H40/40

Abstract: A radiographic imaging apparatus includes a radiation scanner; and a workstation configured to control the radiation scanner. The workstation is configured to output analyzed data for a functional error of at least one of the radiation scanner and the workstation in graphics, and output analyzed data for an item of the functional error in response to an input of a user.

Abstract translation: 放射线照相成像装置包括：放射线扫描器; 以及被配置为控制辐射扫描器的工作站。 工作站被配置为输出分析数据，用于图形中的辐射扫描器和工作站中的至少一个的功能错误，并且响应于用户的输入而输出功能错误项目的分析数据。