Patents by Inventor Tsutomu Horikawa
Tsutomu Horikawa has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).
-
Patent number: 11681910Abstract: Provided are a training apparatus, a recognition apparatus, a training method, a recognition method, and a program that can accurately recognize what an object represented in an image associated with depth information is. An object data acquiring section acquires three-dimensional data representing an object. A training data generating section generates a plurality of training data each representing a mutually different part of the object on the basis of the three-dimensional data. A training section trains a machine learning model using the generated training data as the training data for the object.Type: GrantFiled: July 28, 2017Date of Patent: June 20, 2023Assignee: Sony Interactive Entertainment Inc.Inventors: Tsutomu Horikawa, Daichi Ono
-
Patent number: 11568325Abstract: There are provided a learning apparatus, a learning method, and a program that enable, by using one type of device data, learning of a plurality of models using different data formats. A learning data acquiring section (36) acquires first data that is first-type device data. A first learning section (42) performs learning of a first model (34(1)) in which an estimation using the first-type device data is executed by using the first data. A learning data generating section (40) generates second data that is second-type device data the format of which differs from the format of the first-type device data on the basis of the first data. A second learning section (44) performs learning of a second model (34(2)) in which an estimation using the second-type device data is executed by using the second data.Type: GrantFiled: May 16, 2017Date of Patent: January 31, 2023Assignee: SONY INTERACTIVE ENTERTAINMENT INC.Inventor: Tsutomu Horikawa
-
Publication number: 20220335681Abstract: Provided are an image processing device, an image processing method, and a program capable of forming an accurate three-dimensional map even in a case where a moving object is included in an input image. The image processing device includes an image acquiring section that sequentially acquires two-dimensional captured images, an object type recognition executing section that attaches, to each pixel in the sequentially acquired captured images, a label indicating the type of an object represented by the pixel, and a three-dimensional map creating section that executes three-dimensional position recognition of each pixel of the captured images to create a three-dimensional map, on the basis of the sequentially acquired captured images, and the three-dimensional map creating section restricts the three-dimensional position recognition of each pixel of the captured images according to the label attached to the pixel.Type: ApplicationFiled: September 12, 2019Publication date: October 20, 2022Inventors: Tsutomu HORIKAWA, Daichi ONO, Hiroyuki YABE
-
Publication number: 20220292811Abstract: Provided are an image processing device, an image processing method, and a program for recognizing an object in a three-dimensional map, which do not require collecting learning data of the three-dimensional map and can perform high-speed processing with a small load. The image processing device includes an image acquiring section that sequentially acquires a two-dimensional input image for each frame; an object type recognition executing section that attaches, to each of pixels of the input image acquired for each frame, a label indicating a type of an object represented by the pixels; and a labeling section that executes three-dimensional position recognition of a subject represented in the input image to create a three-dimensional map, based on the input image sequentially input, and attaches, to each voxel included in the three-dimensional map, the label of the pixel corresponding to the voxel.Type: ApplicationFiled: July 12, 2019Publication date: September 15, 2022Inventors: Tsutomu HORIKAWA, Daichi ONO, Hiroyuki YABE
-
Patent number: 11202000Abstract: A learning input image acquisition section acquires a plurality of input images individually depicting a state imaged at a predetermined angle of view in a predetermined relative imaging direction. A learning wide angle-of-view image acquisition section acquires a wide angle-of-view image having an angle of view including all the angles of view of the plurality of input images. A learning section performs learning of a wide angle-of-view image generation section based on a learning wide angle-of-view image and an output that is generated when the plurality of input images are inputted to the wide angle-of-view image generation section, which generates and outputs an image having an angle of view including all the angles of view of a plurality of images in response to the input of the plurality of images.Type: GrantFiled: June 18, 2018Date of Patent: December 14, 2021Assignee: Sony Interactive Entertainment Inc.Inventors: Daichi Ono, Tsutomu Horikawa, Hirotaka Asayama
-
Patent number: 11170246Abstract: Provided are a recognition processing device, a recognition processing method, and a program capable of efficiently narrowing down a three-dimensional region on which recognition processing using a three-dimensional convolutional neural network is to be executed. A first recognition process executing section executes a first recognition process on a captured image obtained by capturing an image of a real space and used to generate voxel data. A target two-dimensional region determining section determines a two-dimensional region occupying part of the captured image on the basis of a result of the first recognition process. A target three-dimensional region determining section determines a three-dimensional region in the real space on the basis of the two-dimensional region and a position of a camera when the camera obtains the captured image.Type: GrantFiled: July 12, 2017Date of Patent: November 9, 2021Assignee: Sony Interactive Entertainment Inc.Inventors: Daichi Ono, Tsutomu Horikawa
-
Patent number: 11164318Abstract: Provided are an image recognition apparatus, an image recognition method, and a program for enabling recognition of many kinds of objects with high precision. An overall recognition unit executes, for at least one given object, a process of recognizing the position of the object in an image. A partial image extraction unit extracts, from the image, a partial image which is a part of the image associated with the recognized position. A partial recognition unit executes a process of recognizing what is one or more objects represented by the partial image, the one or more objects including an object other than the given object the position of which is recognized.Type: GrantFiled: July 18, 2017Date of Patent: November 2, 2021Assignee: Sony Interactive Entertainment Inc.Inventors: Tsutomu Horikawa, Daichi Ono
-
Publication number: 20210218883Abstract: A learning input image acquisition section acquires a plurality of input images individually depicting a state imaged at a predetermined angle of view in a predetermined relative imaging direction. A learning wide angle-of-view image acquisition section acquires a wide angle-of-view image having an angle of view including all the angles of view of the plurality of input images. A learning section performs learning of a wide angle-of-view image generation section based on a learning wide angle-of-view image and an output that is generated when the plurality of input images are inputted to the wide angle-of-view image generation section, which generates and outputs an image having an angle of view including all the angles of view of a plurality of images in response to the input of the plurality of images.Type: ApplicationFiled: June 18, 2018Publication date: July 15, 2021Applicant: Sony Interactive Entertainment Inc.Inventors: Daichi Ono, Tsutomu Horikawa, Hirotaka Asayama
-
Publication number: 20210056337Abstract: Provided are a recognition processing device, a recognition processing method, and a program capable of efficiently narrowing down a three-dimensional region on which recognition processing using a three-dimensional convolutional neural network is to be executed. A first recognition process executing section executes a first recognition process on a captured image obtained by capturing an image of a real space and used to generate voxel data. A target two-dimensional region determining section determines a two-dimensional region occupying part of the captured image on the basis of a result of the first recognition process. A target three-dimensional region determining section determines a three-dimensional region in the real space on the basis of the two-dimensional region and a position of a camera when the camera obtains the captured image.Type: ApplicationFiled: July 12, 2017Publication date: February 25, 2021Applicant: Sony Interactive Entertainment Inc.Inventors: Daichi Ono, Tsutomu Horikawa
-
Publication number: 20200193632Abstract: Provided are a training apparatus, a recognition apparatus, a training method, a recognition method, and a program that can accurately recognize what an object represented in an image associated with depth information is. An object data acquiring section acquires three-dimensional data representing an object. A training data generating section generates a plurality of training data each representing a mutually different part of the object on the basis of the three-dimensional data. A training section trains a machine learning model using the generated training data as the training data for the object.Type: ApplicationFiled: July 28, 2017Publication date: June 18, 2020Applicant: Sony Interactive Entertainment Inc.Inventors: Tsutomu Horikawa, Daichi Ono
-
Publication number: 20200118037Abstract: There are provided a learning apparatus, a learning method, and a program that enable, by using one type of device data, learning of a plurality of models using different data formats. A learning data acquiring section (36) acquires first data that is first-type device data. A first learning section (42) performs learning of a first model (34(1)) in which an estimation using the first-type device data is executed by using the first data. A learning data generating section (40) generates second data that is second-type, device data the format of which differs from the format of the first-type device data on the basis of the first data. A second learning section (44) performs learning of a second model (34(2)) in which an estimation using the second-type device data is executed by using the second data.Type: ApplicationFiled: May 16, 2017Publication date: April 16, 2020Inventor: Tsutomu HORIKAWA
-
Publication number: 20200111215Abstract: Provided are an image recognition apparatus, an image recognition method, and a program for enabling recognition of many kinds of objects with high precision. An overall recognition unit executes, for at least one given object, a process of recognizing the position of the object in an image. A partial image extraction unit extracts, from the image, a partial image which is a part of the image associated with the recognized position. A partial recognition unit executes a process of recognizing what is one or more objects represented by the partial image, the one or more objects including an object other than the given object the position of which is recognized.Type: ApplicationFiled: July 18, 2017Publication date: April 9, 2020Applicant: Sony Interactive Entertainment Inc.Inventors: Tsutomu Horikawa, Daichi Ono
-
Patent number: 8266621Abstract: In an information processing system, when a request for execution of a new task is accepted, one or more of the required processing time and bus bandwidth for the new task are identified. A total required processing time, which is a sum of processing times required for the new task and a task reserved on a scheduler for indicating a procedure pertaining to a plurality of tasks, is compared with a processing time available in an apparatus executing those tasks. Similarly, a total required bus bandwidth, which is a sum of bus bandwidths required for processing the reserved task and the new task, is compared with a bus bandwidth available in the apparatus. Based on the result of the comparisons, it is determined whether the new task is executable.Type: GrantFiled: April 6, 2006Date of Patent: September 11, 2012Assignee: Sony Computer Entertainment Inc.Inventors: Kenichi Murata, Tsutomu Horikawa
-
Patent number: 8266476Abstract: To provide a multiprocessor system in which data transmission efficiency is unlikely to be affected if a damaged processor should exist among a plurality of processors. The multiprocessor system has a plurality of processing modules, including a predetermined number, being three or more, of processors, and a bus for relaying data transmission among the respective processing modules, and specifies at least one damaged processor; selects as a communication restricted processor subjected to communication restriction at least one of the processors connected to the bus at a position determined according to a position where the damaged processor is connected to the bus; and restricts data transmission by the communication restricted processor via the bus.Type: GrantFiled: September 26, 2007Date of Patent: September 11, 2012Assignee: Sony Computer Entertainment Inc.Inventors: Tsutomu Horikawa, Yasukichi Ohkawa
-
Patent number: 8060679Abstract: Requestors acquire tokens before issuing access requests to a memory controller. The access requests issued are accumulated in a command queue of the memory controller. When the amount of access requests accumulated in the command queue is smaller than or equal to a first threshold, or in level 0, tokens are generated at a rate equivalent to 200% of a bus bandwidth. If the amount of accumulation is greater than the first threshold and is smaller than or equal to a second threshold, i.e., in level 1, tokens are generated at a rate equivalent to the bus bandwidth. If the amount of accumulation exceeds the second threshold, the token generation is stopped.Type: GrantFiled: June 11, 2007Date of Patent: November 15, 2011Assignee: Sony Corporation Entertainment Inc.Inventors: Masaaki Nozaki, Tsutomu Horikawa, Kenichi Murata
-
Patent number: 7698473Abstract: Methods and apparatus provide for transferring a plurality of data blocks between a shared memory and a local memory of a processor in response to a single DMA command issued by the processor to a direct memory access controller (DMAC), wherein the processor is capable of operative communication with the shared memory and the DMAC is operatively coupled to the local memory.Type: GrantFiled: January 5, 2005Date of Patent: April 13, 2010Assignee: Sony Computer Entertainment Inc.Inventors: Takeshi Yamazaki, Tsutomu Horikawa, James Allan Kahle, Charles Ray Johns, Michael Norman Day, Peichun Peter Liu
-
Publication number: 20100023665Abstract: To provide a multiprocessor system in which data transmission efficiency is unlikely to be affected if a damaged processor should exist among a plurality of processors. The multiprocessor system has a plurality of processing modules, including a predetermined number, being three or more, of processors, and a bus for relaying data transmission among the respective processing modules, and specifies at least one damaged processor; selects as a communication restricted processor subjected to communication restriction at least one of the processors connected to the bus at a position determined according to a position where the damaged processor is connected to the bus; and restricts data transmission by the communication restricted processor via the bus.Type: ApplicationFiled: September 26, 2007Publication date: January 28, 2010Applicant: SONY COMPUTER ENTERTAINMENT INC.Inventors: Tsutomu Horikawa, Yasukichi Ohkawa
-
Patent number: 7617344Abstract: Requestors issue access requests to a memory controller. The access requests issued are accumulated in a command queue of the memory controller. When the amount of access requests accumulated in the command queue is smaller than or equal to a threshold, a free pass (FP) is granted to specified requesters. When issuing access requests, requesters request and acquire tokens before issuing the access requests if they have no FP granted. If the requesters have an FP, they simply issue the access requests.Type: GrantFiled: June 11, 2007Date of Patent: November 10, 2009Assignee: Sony Computer Entertainment Inc.Inventors: Masaaki Nozaki, Tsutomu Horikawa
-
Publication number: 20080163228Abstract: An information processing method includes: accepting a request for execution of a new task; identifying at least either one of a processing time required for the new task and a bus bandwidth required for processing the new task; making at least either one of comparisons between a total required processing time, which is a sum of processing times required for the new task and a reserved task reserved on a scheduler for indicating a procedure pertaining to a plurality of tasks, and a processing time available in an apparatus that executes those tasks, and between a total required bus bandwidth, which is a sum of bus bandwidths required for processing the reserved task and the new task, and a bus bandwidth available in the apparatus that executes those tasks; determining whether the new task is executable or not, based on the result of comparison(s); making a reservation for the new task on the scheduler if the new task is determined to be executable; and processing the tasks including the new task which is deteType: ApplicationFiled: April 6, 2006Publication date: July 3, 2008Applicant: SONY COMPUTER ENTERTAINMENT INC.Inventors: Kenichi Murata, Tsutomu Horikawa
-
Publication number: 20070294448Abstract: Requestors issue access requests to a memory controller. The access requests issued are accumulated in a command queue of the memory controller. When the amount of access requests accumulated in the command queue is smaller than or equal to a threshold, a free pass (FP) is granted to specified requesters. When issuing access requests, requesters request and acquire tokens before issuing the access requests if they have no FP granted. If the requesters have an FP, they simply issue the access requests.Type: ApplicationFiled: June 11, 2007Publication date: December 20, 2007Applicant: SONY COMPUTER ENTERTAINMENT INC.Inventors: Masaaki Nozaki, Tsutomu Horikawa