公开(公告)号：US12062110B2

公开(公告)日：2024-08-13

申请号：US17250647

申请日：2019-08-16

Applicant: SONY CORPORATION

Inventor： Tooru Masuda

IPC: G06N20/00 ,  G06F18/214 ,  G06T1/20 ,  G06T3/12 ,  G06T3/147 ,  G06T7/73 ,  G06V10/24 ,  G06V10/774

CPC classification number: G06T1/20 ,  G06F18/214 ,  G06N20/00 ,  G06T3/12 ,  G06T3/147 ,  G06T7/73 ,  G06V10/243 ,  G06V10/774 ,  G06T2207/20081 ,  G06T2207/30201

Abstract: An image processing apparatus (100) according to the present disclosure includes: a learning-data creation unit (132) configured to perform projection transformation on image data including a target as a subject, the learning-data creation unit (132) being configured to create learning data including the target as correct data; and a model generation unit (133) configured to generate, based on the learning data created by the learning-data creation unit (132), a learned model for detecting the target included in input data that includes a wide angle view image and is input to the learned model, the wide angle view image being created by projection transformation identical in scheme to the projection transformation by which the learning data is created.

公开(公告)号：US10991144B2

公开(公告)日：2021-04-27

申请号：US16318189

申请日：2017-07-14

Applicant: SONY CORPORATION

Inventor： Yuichi Araki ,  Junichi Tanaka ,  Hiroshi Oryoji ,  Yuichi Hasegawa ,  Tooru Masuda ,  Tetsuya Fukuyasu

IPC: G06T15/04 ,  H04N13/15 ,  G06T15/08 ,  G06T15/10 ,  G06T19/00 ,  H04N13/00

Abstract: There is provided an image processing apparatus that includes an ML3D model generation section, which are applicable to a home server that generates a display image of a predetermined viewpoint from an omnidirectional image or the like. The ML3D model generation section receives transmission information in which auxiliary information is added to at least one of texture information of a first layer, depth information of the first layer, texture information of a second layer or depth information of the second layer, and executes predetermined image processing using the auxiliary information for at least one of the texture information of the first layer, the depth information of the first layer, the texture information of the second layer or the depth information of the second layer.

公开(公告)号：US10898804B2

公开(公告)日：2021-01-26

申请号：US16321525

申请日：2017-08-04

Applicant: SONY CORPORATION

Inventor： Tooru Masuda

IPC: A63F13/525 ,  G06T15/20 ,  G06T19/00 ,  A63F13/655 ,  G06T15/04 ,  G08G1/16 ,  A63F13/5252

Abstract: The present disclosure relates to an image processing device and an image processing method for achieving an easy change of a display mode of each object in live-action content. The image processing device includes an image generation section that changes a display mode of each of objects within a display image on the basis of segment information that indicates a position of a segment in which each of the objects is present, the position of the segment being a position in each of a plurality of layer images that are images generated on the basis of a plurality of captured images and are images classified into a plurality of layers in accordance with distances of the images from a predetermined visual point. For example, the present disclosure is applicable to a display device or the like.

公开(公告)号：US10893212B2

公开(公告)日：2021-01-12

申请号：US16477939

申请日：2018-01-16

Applicant: SONY CORPORATION

Inventor： Tooru Masuda

IPC: H04N5/247 ,  G06T7/55 ,  H04N5/232

Abstract: There is provided an all-celestial imaging apparatus enabling imaging of images that enable estimation of depth information relating to an object to be imaged by suppressing any generation of occlusion. An all-celestial imaging apparatus that is an aspect of the present technique includes plural imaging parts each arranged being directed in a direction different from that of each other, and the plural imaging parts are arranged such that all imaging ranges on at least one circumference of the imaging ranges by the plural imaging parts are each overlapped by angles of view of two or more pairs of the imaging parts. The present technique is applicable to, for example, an all-celestial camera imaging images that are used in the case where the depth information on a distance to an object to be imaged that may be present in an optional direction of all azimuth directions of 360° is estimated.

公开(公告)号：US11354790B2

公开(公告)日：2022-06-07

申请号：US17041270

申请日：2019-01-15

Applicant: SONY CORPORATION

Inventor： Tooru Masuda

IPC: G06T5/00 ,  G06F3/01 ,  G09G5/10

Abstract: An image processing apparatus (100) includes: a variable determination unit (134) that performs, based on a distribution of luminance values of individual pixels included in a predetermined area determined according to information regarding a motion of a user out of first video data recorded in a first dynamic range, determination of a variable to be used to calculate the luminance values of the individual pixels when the first dynamic range is converted into a second dynamic range; and a converter (135) that converts the first video data into second video data displayed in the second dynamic range, based on the variable determined by the variable determination unit.

公开(公告)号：US11132099B2

公开(公告)日：2021-09-28

申请号：US15750953

申请日：2016-06-10

Applicant: Sony Corporation

Inventor： Tooru Masuda

IPC: G06F3/0481 ,  H04N21/431 ,  H04N21/472 ,  H04N21/414 ,  H04N21/81 ,  H04N21/218 ,  H04N5/272 ,  H04N5/45

Abstract: Display of free viewpoint video content is switched so as to alleviate health hazard to viewers without impairing the reality.
The omni window is a window integrating all windows, through each of which each sub-content is looked in from main content, and also serves as an entrance to a free viewpoint space of each sub-content. A viewer can go back and forth between different free viewpoint spaces by way of the omni window and the free viewpoint video is switched in accordance with such an exploring motion within the spaces. Furthermore, it is possible to impart a conversion function for a perceived size of the viewer to the omni window.