Back to results
US 10,474,426 B2Granted
Information processing device, information processing method, and computer program
AssigneeSony Corporation
InventorKenichiro Kobayashi
Life of this patent
Estimating the term…
Abstract
There is provided an information processing device capable of deciding process content of image information according to content of language information input by users, the information processing device including: an image region specifying unit configured to specify a region in an image based on input language information, and a process content specifying unit configured to specify content of a process using the image in regard to the region specified in the image by the image region specifying unit based on the input language information.
Claims19
An information processing device comprising:
- an image region specifying unit configured to
- divide an image captured by a camera into a predetermined number of regions,
- specify a region, of the predetermined number of regions, in the image based on input language information,
- subdivide, based on the input language information indicating a degree, the specified region in the image into a plurality of subdivided regions,
- specify one or more subdivided regions, of the plurality of subdivided regions, in the image based on a level of the degree indicated by the input language information, and
- limit the regions in which a recognition process is performed for an object in the image based on the one or more subdivided regions specified in the image by the image region specifying unit;
- a process content specifying unit configured to specify content of a process using the image in regard to only the one or more subdivided regions specified in the image by the image region specifying unit based on the input language information;
- and
- an information output unit configured to output information regarding the object recognized in the one or more subdivided regions specified in the image by the image region specifying unit,
- wherein a number of the one or more specified subdivided regions, of the plurality of subdivided regions, in the image varies based on the level of the degree,
- wherein the process content specifying unit specifies that a recognition process for the object in only the one or more subdivided regions specified in the image by the image region specifying unit is performed based on the input language information, and
- wherein the image region specifying unit, the process content specifying unit, and the information output unit are each implemented via at least one processor.
9 more elements
An information processing method comprising:
- dividing an image captured by a camera into a predetermined number of regions;
- specifying a region, of the predetermined number of regions, in the image based on input language information;
- subdividing, based on the input language information indicating a degree, the specified region in the image into a plurality of subdivided regions;
- specifying one or more subdivided regions, of the plurality of subdivided regions, in the image based on a level of the degree indicated by the input language information;
- limiting the regions in which an object is recognized in the image based on the one or more subdivided regions specified in the image;
- specifying content of a process using the image in regard to only the one or more specified subdivided regions in the image based on the input language information;
- specifying that a recognition process for the object in only the one or more specified subdivided regions is performed based on the input language information;
- and
- outputting information regarding the object recognized in the one or more subdivided regions specified in the image,
- wherein a number of the one or more specified subdivided regions, of the plurality of subdivided regions, in the image varies based on the level of the degree.
7 more elements
A non-transitory computer-readable medium having embodied thereon a program, which when executed by a computer causes the computer to execute a method, the method comprising:
- dividing an image captured by a camera into a predetermined number of regions;
- specifying a region, of the predetermined number of regions, in the image based on input language information;
- subdividing, based on the input language information indicating a degree, the specified region in the image into a plurality of subdivided regions;
- specifying one or more subdivided regions, of the plurality of subdivided regions, in the image based on a level of the degree indicated by the input language information;
- limiting the regions in which an object is recognized in the image based on the one or more subdivided regions specified in the image;
- specifying content of a process using the image in regard to only the one or more specified subdivided regions in the image based on the input language information;
- specifying that a recognition process for the object in only the one or more specified subdivided regions is performed based on the input language information;
- and
- outputting information regarding the object recognized in the one or more subdivided regions specified in the image,
- wherein a number of the one or more specified subdivided regions, of the plurality of subdivided regions, in the image varies based on the level of the degree.
7 more elements
Dependencies are inferred from the claim text.