-
公开(公告)号：US20210104035A1
公开(公告)日：2021-04-08
申请号：US17126655
申请日：2020-12-18
Inventor： Yuki MARUYAMA , Hiroya KUSAKA , Taro IMAGAWA , Akihiro NODA
Abstract: An inspection device includes: an obtainer that obtains at least one captured image of a plurality of sound insulation plates; a deriver that derives a displacement of each of the plurality of sound insulation plates in the image; and an extractor that, based on the displacements of the plurality of sound insulation plates, extracts, from the plurality of sound insulation plates, a sound insulation plate of interest that moves differently from another sound insulation plate included in the plurality of sound insulation plates.
-
公开(公告)号：US20210104032A1
公开(公告)日：2021-04-08
申请号：US17126727
申请日：2020-12-18
Inventor： Akihiro NODA , Taro IMAGAWA , Hiroya KUSAKA , Yuki MARUYAMA
Abstract: A displacement measurement device includes: a first machine learning model trained to generate, from one image which contains a subject and has noise, at least one image which contains the subject and which has noise or has had noise removed; a first obtainer that obtains a first image which contains the subject and has noise and a second image which contains the subject and has noise; a first generator that, using the first machine learning model, generates M template images containing the subject from the first image and generates M target images containing the subject from the second image, M being an integer of 2 or higher; a hypothetical displacement calculator that calculates M hypothetical displacements of the subject from the M template images and the M target images; and a displacement calculator that calculates a displacement of the subject by performing statistical processing on the M hypothetical displacements.
-
公开(公告)号：US20190289081A1
公开(公告)日：2019-09-19
申请号：US16431017
申请日：2019-06-04
Inventor： Youji SHIBAHARA , Toshiyasu SUGIO , Toru MATSUNOBU , Yuki MARUYAMA , Takahiro NISHI
Abstract: A method controls access permission to an image including receiving session information from a first camera at a first position, the session information indicating a subject to be photographed and a session in which the first camera joins, receiving an image of the subject from the first camera, updating the session information such that a second camera, at a second position, joins the session, and monitoring whether the second camera joins the session. A first period is calculated in which the second camera joins the session, a second period is calculated based on the first period, and access permission to a shared image is controlled such that the second camera downloads the shared image during the second period and does not download the shared image outside of the second period, the shared image being one of the image and an image generated based on the image.
-
公开(公告)号：US20170251226A1
公开(公告)日：2017-08-31
申请号：US15293759
申请日：2016-10-14
Inventor： Toshihiko KUSAKABE , Kiyofumi ABE , Hideyuki OHGOSE , Shinji KITAMURA , Takashi HASHIMOTO , Yuki MARUYAMA , Tatsuro JURI
CPC classification number: H04N19/436 , H04N19/70
Abstract: The image encoding apparatus which encodes, on a per block basis, a current picture included in a moving picture includes: an encoder which outputs a first encoded stream including, in the following arrangement order, (i) slice data obtained by encoding a slice included in the current picture and including a plurality of block lines and (ii) a slice header including information indicating a code amount of each of the plurality of block lines; and an exchanger which exchanges arrangement positions of the slice data and the slice header in the first encoded stream.
-
Patent Agency Ranking
15.发明公开公开(公告)号：US20230252650A1
公开(公告)日：2023-08-10
申请号：US18134256
申请日：2023-04-13
Inventor： Taro IMAGAWA , Yuki MARUYAMA , Hiroya KUSAKA
IPC: G06T7/246
CPC classification number: G06T7/248 , G06T2207/30184
Abstract: An information processing method includes: obtaining training images generated by a camera shooting a target point and training parameters related to a position/orientation of the camera when shooting the training images; generating, using the training parameters as inputs, a model that outputs a first displacement amount by which a target block corresponding to the target point displaces from a reference position in the training images; obtaining measurement images generated by the camera shooting the target point and measurement parameters related to the position/orientation of the camera when shooting the measurement images; obtaining a second displacement amount, output by inputting the measurement parameters into the model, by which a target block corresponding to the target point appearing in the measurement images displaces from a reference position in the measurement images; and outputting a displacement obtained by subtracting the second displacement amount from displacement of the target block in the measurement images.
-
公开(公告)号：US20220128433A1
公开(公告)日：2022-04-28
申请号：US17568129
申请日：2022-01-04
Inventor： Taro IMAGAWA , Akihiro NODA , Yuki MARUYAMA , Hiroya KUSAKA
Abstract: An imaging parameter output method is a method of outputting an imaging parameter of an imaging device that captures an image for measuring a displacement representing a movement of an object. The imaging parameter output method includes: obtaining object information identifying the object, and a geometric imaging condition for imaging the object; calculating the imaging parameter including a candidate imaging area for placing the imaging device and the accuracy in measuring the displacement in the candidate imaging area, based on the object information and the geometric imaging condition, without imaging the object using the imaging device; and outputting the imaging parameter.
-
公开(公告)号：US20210110515A1
公开(公告)日：2021-04-15
申请号：US17132163
申请日：2020-12-23
Inventor： Akihiro NODA , Taro IMAGAWA , Yuki MARUYAMA , Hiroya KUSAKA
Abstract: A displacement measurement device includes: an obtainer that obtains a first image which contains a subject and a second image which contains the subject; a generator that generates M template images which contain the subject and which have noise from the first image and generates M target images which contain the subject and which have noise from the second image, M being an integer of 2 or higher; a hypothetical displacement calculator that calculates M hypothetical displacements of the subject from the M template images and the M target images; and a displacement calculator that calculates a displacement of the subject by performing statistical processing on the M hypothetical displacements.
-
公开(公告)号：US20170251215A1
公开(公告)日：2017-08-31
申请号：US15298603
申请日：2016-10-20
Inventor： Kiyofumi ABE , Hideyuki OHGOSE , Shinji KITAMURA , Takashi HASHIMOTO , Toshihiko KUSAKABE , Yuki MARUYAMA , Tatsuro JURI
IPC: H04N19/174 , H04N19/159 , H04N19/115 , H04N19/196
CPC classification number: H04N19/174 , H04N19/105 , H04N19/115 , H04N19/159 , H04N19/16 , H04N19/172 , H04N19/196
Abstract: A moving picture encoding apparatus encodes a moving picture having an interlaced structure, and includes: a storage which stores fields as reference pictures; and an encoder which encodes a current field as a B-picture, using a first reference picture list which includes only one field in a same parity as the current field, and a second reference picture list which includes only one field in an opposite parity to the current field.
-
公开(公告)号：US20160021374A1
公开(公告)日：2016-01-21
申请号：US14800643
申请日：2015-07-15
Inventor： Yuki MARUYAMA , Kiyofumi ABE , Hideyuki OHGOSE , Koji ARIMURA , Hiroshi ARAKAWA , Kazuma SAKAKIBARA , Kazuhito KIMURA
IPC: H04N19/136 , H04N19/593 , H04N19/176
CPC classification number: H04N19/159 , H04N19/11 , H04N19/186
Abstract: Provided is an image coding method which obtains a picture, and codes the obtained picture. The image coding method generates a processing block, performs a first intra prediction that selects one intra prediction mode from a first set of candidates based on a component of a first signal contained in the processing block, performs a second intra prediction that selects one intra prediction mode from a second set of candidates based on a component of a second signal contained in the processing block. The first set of candidates includes a plurality of intra prediction modes, and the second set of candidates includes an intra prediction mode, having no dependency between the second intra prediction and the first intra prediction, of the plurality of intra prediction modes. The first intra prediction and the second intra prediction are performed in parallel.
Abstract translation: 提供一种图像编码方法，其获得图像，并对获得的图像进行编码。 图像编码方法根据包含在处理块中的第一信号的分量，生成处理块，执行从第一候选集合中选择一个帧内预测模式的第一帧内预测，执行选择一个帧内预测的第二帧内预测 模式从基于包含在处理块中的第二信号的分量的第二组候选。 第一候选项包括多个帧内预测模式，并且第二候选组包括多个帧内预测模式中不具有第二帧内预测和第一帧内预测之间的依赖关系的帧内预测模式。 并行执行第一帧内预测和第二帧内预测。
-
-
-
-
-
-
-
-
Search Results
19
records
(took 0.02 seconds)
