公开(公告)号：KR20070100160A

公开(公告)日：2007-10-10

申请号：KR20070033807

申请日：2007-04-05

Applicant: SONY CORP

Inventor： MINAMINO KATSUKI ,  ITO MASATO ,  KAWAMOTO KENTA ,  YOSHIIKE YUKIKO ,  SUZUKI HIROTAKA

IPC: G06M3/08 ,  G06F15/18 ,  G06M3/06

CPC classification number: G06N99/005 ,  G06N3/0445 ,  G06N3/0454 ,  G06N3/084

Abstract: A learning apparatus, a learning method, and a program are provided to learn the dynamics of a network in a self organizing manner according to the value corresponding to the learning weighting value through a learning unit. A learning apparatus comprises a memory unit, a learning unit(1-4), a winner node determining unit(7-2), and a weighting value determining unit(7-3). The memory unit memorizes the network formed by plural nodes maintaining the dynamics. The learning unit learns the dynamics of the network in a self-organizing manner on the basis of the time-series data. The winner node determining unit determines a winner node which is the node having the dynamics proper to the time-series data. The weighting value determining unit determines the weighting value of the dynamics according to the distance of the individual node from the winner node.

Abstract translation: 提供学习装置，学习方法和程序，以通过学习单元根据与学习加权值对应的值以自组织方式学习网络的动态。 学习装置包括存储单元，学习单元（1-4），胜利者节点确定单元（7-2）和加权值确定单元（7-3）。 存储单元存储由维持动态的多个节点形成的网络。 学习单元根据时间序列数据以自组织的方式学习网络的动态。 优胜者节点确定单元确定胜利者节点，其是具有适合于时间序列数据的动力的节点。 加权值确定单元根据个体节点与获胜者节点的距离来确定动态的权重值。

公开(公告)号：WO2005088244A9

公开(公告)日：2008-03-13

申请号：PCT/JP2005004839

申请日：2005-03-17

Applicant: SONY CORP ,  GUTMANN STEFFEN ,  FUKUCHI MASAKI ,  SABE KOHTARO ,  KAWAMOTO KENTA

Inventor： GUTMANN STEFFEN ,  FUKUCHI MASAKI ,  SABE KOHTARO ,  KAWAMOTO KENTA

IPC: G01B11/24 ,  B62D57/024 ,  G06K9/00 ,  G06T1/00 ,  G06T7/00

CPC classification number: G06K9/00664 ,  B62D57/024 ,  G06K9/00201 ,  G06T7/13 ,  G06T7/73 ,  G06T2207/30252

Abstract: A plane detecting unit (3) of a plane detector has a line segment extracting section (4) which selects a group of distance data points on the same plane from among distance data points forming an image and extracts line segments from the distance data point group and a region expanding section (5) which detects one or more plane regions present in the image from a line segment group of all the line segments extracted by the line segment extracting section (4). The line segment extracting section (4) draws a line segment (L1) connecting the ends of the distance data point group, seeks a point of interest (brk) from which the distance to the line segment (L1) is the largest, divides the data point group at the point of interest if the distance is a predetermined value or more, determines a line segment (L2) by the least squares method if the distance is below the predetermined value, judges that the line segment is of a zigzag shape if a predetermined number of data points are continuously present on one side of the line segment (L2), divides the data point group at the point of interest (brk), and repeats the above processings. With this, planes are correctly detected at a time from the distance data including measurement noise robustly against noise.

Abstract translation: 平面检测器的平面检测单元（3）具有线段提取部（4），其从形成图像的距离数据点中选择同一平面上的一组距离数据点，并从距离数据点组提取线段 以及区域扩展部（5），其从由所述线段提取部（4）提取的所有线段的线段组检测所述图像中存在的一个以上的平面区域。 线段提取部分（4）绘制连接距离数据点组的端部的线段（L1），寻找距离线段（L1）的距离最大的兴趣点（brk） 如果距离是预定值或更大，则在感兴趣点处的数据点组，如果距离低于预定值，则通过最小二乘法确定线段（L2），判断线段是否为之字形，如果 在线段（L2）的一侧上连续存在预定数量的数据点，在兴趣点（brk）处划分数据点组，并重复上述处理。 由此，可以从包括测量噪声的距离数据一次正确地检测到平面，以抵抗噪声。

公开(公告)号：DE60216411D1

公开(公告)日：2007-01-11

申请号：DE60216411

申请日：2002-08-21

Applicant: SONY CORP

Inventor： YOKONO JUN ,  SABE KOHTARO ,  KAWAMOTO KENTA

IPC: G06K9/00 ,  G06K9/62

Abstract: A robot includes a face extracting section for extracting features of a face included in an image captured by a CCD camera, and a face recognition section for recognizing the face based on a result of face extraction by the face extracting section. The face extracting section is implemented by Gabor filters that filter images using a plurality of filters that have orientation selectivity and that are associated with different frequency components. The face recognition section is implemented by a support vector machine that maps the result of face recognition to a non-linear space and that obtains a hyperplane that separates in that space to discriminate a face from a non-face. The robot is allowed to recognize a face of a user within a predetermined time under a dynamically changing environment.

公开(公告)号：DE602004004048D1

公开(公告)日：2007-02-15

申请号：DE602004004048

申请日：2004-11-22

Applicant: SONY CORP

Inventor： HIDAI KENICHI ,  SABE KOHTARO ,  KAWAMOTO KENTA

IPC: G06K9/62 ,  G06T1/00 ,  G06K9/00 ,  G06K9/68 ,  G06N3/08 ,  G06T7/00

Abstract: An object detecting device 1 comprises a scaling section 3 for generating scaled images by scaling down a gradation image input from an image output section 2, a scanning section 4 for sequentially manipulating the scaled images and cutting out window images from them and a discriminator 5 for judging if each window image is an object or not. The discriminator 5 includes a plurality of weak discriminators that are learnt in a group by boosting and an adder for making a weighted majority decision from the outputs of the weak discriminators. Each of the weak discriminators outputs an estimate telling the likelihood of a window image to be an object or not by using the difference of the luminance values of two pixels. The discriminator 5 suspends the operation of computing estimates for a window image that is judged to be a non-object, using a threshold value that is learnt in advance.

公开(公告)号：DE602004004048T2

公开(公告)日：2007-07-12

申请号：DE602004004048

申请日：2004-11-22

Applicant: SONY CORP

Inventor： HIDAI KENICHI ,  SABE KOHTARO ,  KAWAMOTO KENTA

IPC: G06K9/62 ,  G06T1/00 ,  G06K9/00 ,  G06K9/68 ,  G06N3/08 ,  G06T7/00

Abstract: An object detecting device 1 comprises a scaling section 3 for generating scaled images by scaling down a gradation image input from an image output section 2, a scanning section 4 for sequentially manipulating the scaled images and cutting out window images from them and a discriminator 5 for judging if each window image is an object or not. The discriminator 5 includes a plurality of weak discriminators that are learnt in a group by boosting and an adder for making a weighted majority decision from the outputs of the weak discriminators. Each of the weak discriminators outputs an estimate telling the likelihood of a window image to be an object or not by using the difference of the luminance values of two pixels. The discriminator 5 suspends the operation of computing estimates for a window image that is judged to be a non-object, using a threshold value that is learnt in advance.

公开(公告)号：DE602004030893D1

公开(公告)日：2011-02-17

申请号：DE602004030893

申请日：2004-03-24

Applicant: SONY CORP

Inventor： SABE KOHTARO ,  OHASHI TAKESHI ,  KAWAMOTO KENTA

IPC: B25J13/00 ,  B25J5/00 ,  B25J13/08 ,  B62D57/032

公开(公告)号：DE60216411T2

公开(公告)日：2007-10-04

申请号：DE60216411

申请日：2002-08-21

Applicant: SONY CORP

Inventor： YOKONO JUN ,  SABE KOHTARO ,  KAWAMOTO KENTA

IPC: G06K9/00 ,  G06K9/62

Abstract: A robot includes a face extracting section for extracting features of a face included in an image captured by a CCD camera, and a face recognition section for recognizing the face based on a result of face extraction by the face extracting section. The face extracting section is implemented by Gabor filters that filter images using a plurality of filters that have orientation selectivity and that are associated with different frequency components. The face recognition section is implemented by a support vector machine that maps the result of face recognition to a non-linear space and that obtains a hyperplane that separates in that space to discriminate a face from a non-face. The robot is allowed to recognize a face of a user within a predetermined time under a dynamically changing environment.

公开(公告)号：EP2985652A4

公开(公告)日：2016-12-14

申请号：EP14783184

申请日：2014-03-26

Applicant: SONY CORP

Inventor： TANAKA EIICHI ,  TATSUTA HIROKAZU ,  HASEGAWA SHIN ,  HASEGAWA YUICHI ,  KAWAMOTO KENTA

IPC: G02B27/02 ,  H04N5/64

CPC classification number: G02B27/0176 ,  G02B9/16 ,  G02B13/06 ,  G02B27/0172 ,  G02B2027/0123 ,  G02B2027/0132 ,  G02B2027/0154 ,  G02B2027/0161 ,  G02B2027/0169

公开(公告)号：EP1343115A4

公开(公告)日：2005-03-09

申请号：EP02762820

申请日：2002-08-21

Applicant: SONY CORP

Inventor： YOKONO JUN ,  SABE KOHTARO ,  KAWAMOTO KENTA

IPC: G06K9/00 ,  G06K9/62

CPC classification number: G06K9/00288 ,  G06K9/00221 ,  G06K9/00228

Abstract: A robot includes a face extraction unit for extracting a feature of a face contained in an image picked up by a CCD camera and a face recognition unit for recognizing a face according to the face extraction result obtained by the face extraction unit. The face extraction unit is composed of a Gabor filter for filtering an image by using a plurality of filters having direction selectivity and different frequency components. The face recognition unit is composed of a support vector machine for mapping the face extraction result onto a non-linear space and obtaining a hyperplane for separation in the space, thereby distinguishing face from non-face. The robot can recognize a user face in a dynamically changing environment within a predetermined time.

公开(公告)号：EP3023829A4

公开(公告)日：2017-05-10

申请号：EP14825860

申请日：2014-05-16

Applicant: SONY CORP

Inventor： TANAKA EIICHI ,  TATSUTA HIROKAZU ,  HASEGAWA SHIN ,  HASEGAWA YUICHI ,  KAWAMOTO KENTA

IPC: G02B27/01 ,  G02F1/1333 ,  H01L51/00

CPC classification number: G02B27/0172 ,  G02B17/026 ,  G02B27/0176 ,  G02B2027/0123 ,  G02B2027/0129 ,  G02B2027/0132 ,  G02B2027/0159 ,  G02B2027/0161 ,  G02B2027/0169 ,  H01L51/0097 ,  H01L2251/5338 ,  Y02E10/549

Abstract: A display apparatus includes a frame and an image display apparatus for a left eye 3 and an image display apparatus for a right eye which are attached to the frame. Each image display apparatus 30 includes an image forming apparatus 40 and an optical system 50 configured to guide an image from the image forming apparatus 40 to a pupil of an observer 10. The optical system 50 includes a reflecting mirror 51 and a lens group 52. The image forming apparatus 40 is curved. A normal NLL of the reflecting mirror 51 included in the optical system of the image display apparatus for a left eye and a normal NLR of the reflecting mirror 51 included in the optical system of the image display apparatus for a right eye intersect with each other in a space on an opposite side of the observer 10 with respect to the reflecting mirror 51.

Abstract translation: 显示设备包括框架和附接到框架的用于左眼的图像显示设备3和用于右眼的图像显示设备。 各图像显示装置30具备图像形成装置40和将来自图像形成装置40的图像引导至观察者10的瞳孔的光学系统50.光学系统50具有反射镜51和透镜组52。 图像形成装置40是弯曲的。 包含在左眼用图像显示装置的光学系统中的反射镜51的法线NLL和右眼用图像显示装置的光学系统中包含的反射镜51的法线NLR相互交叉 观察者10相对于反射镜51的相对侧上的空间。