公开(公告)号：US11100352B2

公开(公告)日：2021-08-24

申请号：US16576278

申请日：2019-09-19

Applicant: Samsung Electronics Co., Ltd.

Inventor： Hyunku Lee ,  Hyunsurk Ryu ,  Ha Young Kim ,  BaRom Kang

IPC: G06K9/32 ,  G06F17/15 ,  G06N3/08 ,  G06T7/70

Abstract: Disclosed is a computer-readable medium including a program code that, when executed by processing circuitry, causes the processing circuitry to generate a feature map from an input image, to extract a region of interest from the feature map, and to generate a predicted mask based on the region of interest. The processing circuitry may use a predicted mask and a real mask to learn a convolutional neural network system. The real mask includes first pixels corresponding to the real boundary and second pixels corresponding to a fake boundary adjacent to the real boundary.

公开(公告)号：US10531031B2

公开(公告)日：2020-01-07

申请号：US15914073

申请日：2018-03-07

Applicant: Samsung Electronics Co., Ltd.

Inventor： Keun Joo Park ,  Junseok Kim ,  Yohan Roh ,  Hyunsurk Ryu ,  Jooyeon Woo ,  Hyunku Lee

IPC: H04N5/3745 ,  H04N5/341 ,  H05B39/04 ,  G06F16/23 ,  G06F16/583 ,  H04N5/367 ,  H05B37/02

Abstract: An image processing device includes a vision sensor and a processor. The vision sensor generates a plurality of events in which an intensity of light changes and generates a plurality of timestamps depending on times when the events occur. In addition, the processor may regenerate a timestamp of a pixel where an abnormal event occurs, based on temporal correlation of the events.

公开(公告)号：US20190065885A1

公开(公告)日：2019-02-28

申请号：US16113409

申请日：2018-08-27

Applicant: SAMSUNG ELECTRONICS CO., LTD.

Inventor： Jia Li ,  Feng Shi ,  Weiheng Liu ,  Dongqing Zou ,  Qiang Wang ,  Hyunsurk Ryu ,  Keun Joo Park ,  Hyunku Lee

IPC: G06K9/46 ,  G06T7/246 ,  G06K9/00 ,  H04N7/18

Abstract: An object detection method and a neural network system for object detection are disclosed. The object detection method acquires a current frame of a sequence of frames representing an image sequence, and extracts a feature map of the current frame. The extracted feature map is pooled with information of a pooled feature map of a previous frame to thereby obtain a pooled feature map of the current frame. An object is detected from the pooled feature map of the current frame. A dynamic vision sensor (DVS) may be utilized to provide the sequence of frames. Improved object detection accuracy may be realized, particularly when object movement speed is slow.

公开(公告)号：US09317127B2

公开(公告)日：2016-04-19

申请号：US14199164

申请日：2014-03-06

Applicant: SAMSUNG ELECTRONICS CO., LTD. ,  UNIVERSITY OF ZURICH

Inventor： Jun Haeng Lee ,  Tobi Delbruck ,  Hyunsurk Ryu ,  Keun Joo Park ,  Chang Woo Shin

IPC: G06F3/01 ,  G06K9/62 ,  G06K9/00 ,  G06T7/20

CPC classification number: G06F3/017 ,  G06K9/00335 ,  G06T7/254 ,  G06T2200/28

Abstract: A motion recognizing apparatus and method are provided. According to an aspect, a motion recognizing apparatus may include: an optical sensor configured to sense at least a portion of a subject where a motion occurs and to output one or more events in response thereto; a motion tracing unit configured to trace a motion locus of the portion where the motion occurs based on the one or more outputted events; and a motion pattern determining unit configured to determine a motion pattern of the portion where the motion occurs based on the traced motion locus.

Abstract translation: 提供了一种运动识别装置和方法。 根据一方面，运动识别装置可以包括：光学传感器，被配置为感测发生运动的对象的至少一部分并且响应于此而输出一个或多个事件; 运动跟踪单元，被配置为基于所述一个或多个输出事件跟踪所述运动发生的部分的运动轨迹; 以及运动图案确定单元，被配置为基于跟踪的运动轨迹来确定运动发生的部分的运动模式。