公开(公告)号：US20240119605A1

公开(公告)日：2024-04-11

申请号：US18275786

申请日：2021-02-09

Applicant: NIPPON TELEGRAPH AND TELEPHONE CORPORATION

Inventor： Ken NAKAMURA ,  Hiroyuki UZAWA ,  Daisuke KOBAYASHI ,  Saki HATTA ,  Yuya OMORI ,  Shuhei YOSHIDA

IPC: G06T7/215 ,  G06T7/246 ,  G06V10/82 ,  G06V20/40

CPC classification number: G06T7/215 ,  G06T7/248 ,  G06V10/82 ,  G06V20/46 ,  G06T2207/10016 ,  G06T2207/20084

Abstract: A calculation unit (22) sets each frame, of a moving image including a plurality of frames, as a target frame, calculates a motion vector with reference to a reference frame over an entirety of the target frames, and calculates an index indicating a magnitude of change between a key frame and the target frame, which is represented by using the motion vector, a judgement unit (25) judges whether or not the calculated index is equal to or greater than a predetermined threshold value, a first detection unit (27) detects a region indicating an object from the target frame by using an object detection model in a case in which the index is equal to greater than the threshold value, a second detection unit (28) detects a region on the target frame, which is obtained by correcting a position of a region detected in the reference frame by using the motion vector in a case in which the index is less than the threshold value, and an output unit (30) outputs information of the region detected by the first detection unit (27) and information of the region detected by the second detection unit (28).

公开(公告)号：US20240114130A1

公开(公告)日：2024-04-04

申请号：US18277066

申请日：2021-02-16

Applicant: NIPPON TELEGRAPH AND TELEPHONE CORPORATION

Inventor： Daisuke KOBAYASHI ,  Ken NAKAMURA ,  Koyo NITTA ,  Yuya OMORI

IPC: H04N19/11 ,  H04N19/119 ,  H04N19/176 ,  H04N19/593

CPC classification number: H04N19/11 ,  H04N19/119 ,  H04N19/176 ,  H04N19/593

Abstract: A video coding method is a video coding method for coding an image in units of blocks obtained by dividing an image, and in the method a computer performs processing for dividing the image into blocks, acquiring an intra-image prediction mode predicted for each block, and generating, by using a combination pattern based on the intra-image prediction mode of a plurality of blocks adjacent to at least any one block, determining whether or not the plurality of blocks are to be combined, and performing prediction for each combined block when it is determined that the plurality of blocks are to be combined.

公开(公告)号：US20240370694A1

公开(公告)日：2024-11-07

申请号：US18561276

申请日：2021-05-26

Applicant: NIPPON TELEGRAPH AND TELEPHONE CORPORATION

Inventor： Saki HATTA ,  Hiroyuki UZAWA ,  Shuhei YOSHIDA ,  Daisuke KOBAYASHI ,  Yuya OMORI ,  Ken NAKAMURA ,  Koyo NITTA

IPC: G06N3/04

Abstract: An object detection device subjects fixed-length data having a decimal point position set therein to an arithmetic processing corresponding to respective layers in a plurality of layers configuring a multilayer neural network to which an input image is input, the arithmetic processing being performed in accordance with a processing algorithm for the multilayer neural network to which an input image is input. In the arithmetic processing, the object detection device counts the upper limit number of saturations, which is a number of times that upper limit value of a value range determined by the decimal point position is exceeded, and the lower limit number of saturations, which is a number of times that the lower limit value of the value range is not reached. The object detection device counts the upper limit number of saturation layers, which is a number of layers in which the upper limit number of saturations is one or larger, and the lower limit number of saturation layers, which is a number of layers in which the lower limit number of saturations is one or larger. The object detection device changes at least one of the upper limit saturation threshold, which is the threshold of the upper limit number of saturations or the lower limit saturation threshold, which is the threshold of the lower limit number of saturations, when at least one of the upper limit saturation threshold or the lower limit saturation threshold is determined not to be optimal based on an amount of change in the upper limit number of saturation layers and an amount of change in the lower limit number of saturation layers. The object detection device sets the decimal point position for each layer in the plurality of layers, based on a result of the determination.

公开(公告)号：US20240062506A1

公开(公告)日：2024-02-22

申请号：US18265881

申请日：2020-12-09

Applicant: NIPPON TELEGRAPH AND TELEPHONE CORPORATION

Inventor： Hiroyuki UZAWA ,  Saki HATTA ,  Shuhei YOSHIDA ,  Daisuke KOBAYASHI ,  Yuya OMORI ,  Ken NAKAMURA ,  Koyo NITTA

IPC: G06V10/26 ,  G06T3/40 ,  G06V10/22

CPC classification number: G06V10/267 ,  G06T3/40 ,  G06V10/225 ,  G06V10/273 ,  G06V2201/10

Abstract: An object detection device 10 includes an entire processing unit 110 that obtains first metadata for the entire input image by scaling the input image and performing object detection processing, a divided image narrowing unit 120 that narrows down the input image into a predetermined number of selected divided images from a group of divided images obtained by dividing the input image, a division processing unit 130 that obtains second metadata by performing object detection processing for each of the selected divided images, and a synthesis processing unit 140 that removes the second metadata obtained by the division processing unit 130 that overlaps the first metadata obtained by the entire processing unit 110, and synthesizes the first metadata not removed and the first meta data obtained by the entire processing unit 110 to output the meta data.

公开(公告)号：US20240135155A1

公开(公告)日：2024-04-25

申请号：US18269528

申请日：2020-12-28

Applicant: NIPPON TELEGRAPH AND TELEPHONE CORPORATION

Inventor： Saki HATTA ,  Hiroyuki UZAWA ,  Shuhei YOSHIDA ,  Daisuke KOBAYASHI ,  Yuya OMORI ,  Ken NAKAMURA ,  Koyo NITTA

IPC: G06N3/063

CPC classification number: G06N3/063

Abstract: In a data processing device, a fixed-point position control unit determines, as first control. The fixed-point position control unit causes a detection calculation unit to perform calculation processing on processing target data at a processing point in time. The saturation rate control unit instructs, as second control to be repeated by the fixed-point position control unit, the fixed-point position control unit to move at least the fixed-point position as control to increase a lower limit saturation rate proportional to a magnitude of a counted lower limit counter value with respect to a result of the first control. The fixed-point position control unit performs, as the second control, a predetermined determination on the basis of the instruction from the saturation rate control unit and the metadata, determines the fixed-point position moved for each layer, and causes calculation processing to be performed.

公开(公告)号：US20240054181A1

公开(公告)日：2024-02-15

申请号：US18256005

申请日：2020-12-09

Applicant: NIPPON TELEGRAPH AND TELEPHONE CORPORATION

Inventor： Yuya OMORI ,  Ken NAKAMURA ,  Daisuke KOBAYASHI ,  Koyo NITTA

IPC: G06F17/16 ,  G06F7/50

CPC classification number: G06F17/16 ,  G06F7/50

Abstract: One aspect of the present invention is an operation circuit for performing a convolution operation of input feature map information supplied as a plurality of channels and coefficient information supplied as a plurality of channels, the operation circuit including a set including at least two channels of an output feature map based on output channels and at least three sub-operation circuits, wherein at least two sub-operation circuits are allocated for each set, the sub-operation circuits included in the set execute processing of a convolution operation of the coefficient information and the input feature map information included in the set, when a specific channel of the output feature map is a zero matrix, a sub-operation circuit that performs a convolution operation of the zero matrix executes processing of a convolution operation of the coefficient information and the input feature map information to be supplied next from a channel of the output feature map and a channel of the input feature map included in the set, and a result of the convolution operation is output for each channel of the output feature map.

公开(公告)号：US20210258591A1

公开(公告)日：2021-08-19

申请号：US16973741

申请日：2019-06-04

Applicant: NIPPON TELEGRAPH AND TELEPHONE CORPORATION

Inventor： Yuya OMORI ,  Takayuki ONISHI ,  Hiroe IWASAKI ,  Atsushi SHIMIZU

IPC: H04N19/423 ,  G11B20/00 ,  G11B20/10 ,  H04N19/433

Abstract: A buffer device stores, in relation to a first area and a second area, which are partial areas respectively located on an identical line of an encoding target frame, and a third area, which is a partial area on the line below the identical line, reference image data for each search range of a reference frame that is referred to during motion search processing executed in order of the first area, the second area, and the third area, the buffer device including a memory for storing the reference image data, and a buffer control unit which, when the motion search processing is executed on the second area, executes control so that among the reference image data included in a first search range that is referred to during the motion search processing of the first area, the reference image data included in a third search range that is referred to during the motion search processing of the third area are left in the memory without being deleted.

公开(公告)号：US20250036715A1

公开(公告)日：2025-01-30

申请号：US18715083

申请日：2021-12-03

Applicant: NIPPON TELEGRAPH AND TELEPHONE CORPORATION

Inventor： Saki HATTA ,  Hiroyuki UZAWA ,  Shuhei YOSHIDA ,  Yuko IINUMA ,  Yuya OMORI ,  Daisuke KOBAYASHI ,  Ken NAKAMURA

IPC: G06F17/15

Abstract: There is provided a data processing device 10 that performs a convolution operation of two pieces of input data of 2M×N bits (N is a positive integer and M is a natural number) width with a minimum accuracy of the convolution operation being N bits, and performs processing corresponding to a plurality of the consecutive M, the data processing device 10 including: a product-sum operation unit 101 that performs a product-sum operation according to the value of M; a shifter 102 that performs shift processing on a result of a product-sum operation of the product-sum operation unit 101 in a case where the value of M is not 0; an addition unit 103 that performs addition processing on each output of the shifter 102 or the product-sum operation unit 101 according to the value of M; a selector 105 that selects an output from the addition unit 103 according to the value of M; a cumulative addition unit 106 that cumulatively adds the outputs from the selector 105; and a cumulative addition memory 107 that stores outputs from the cumulative addition unit 106 in a process of a convolution operation.

公开(公告)号：US20240289593A1

公开(公告)日：2024-08-29

申请号：US18572329

申请日：2021-06-25

Applicant: NIPPON TELEGRAPH AND TELEPHONE CORPORATION

Inventor： Shuhei YOSHIDA ,  Hiroyuki UZAWA ,  Daisuke KOBAYASHI ,  Ken NAKAMURA ,  Saki HATTA ,  Yuya OMORI

IPC: G06N3/0464

CPC classification number: G06N3/0464

Abstract: A first aspect of the present disclosure is a convolutional neural network inference processing device that performs processing in a convolutional neural network including a plurality of convolution layers and a residual layer that adds intermediate data related to the plurality of convolution layers as an addition target to a processing result by the plurality of convolution layers for each tile that is data obtained by dividing input data into a predetermined size, the convolutional neural network inference processing device including an inconsistency data storage unit that stores inconsistency data that is data at a portion where there is inconsistency between the processing result and the intermediate data, a past layer data storage unit that stores past layer data that is an addition target in a residual layer generated using inconsistency data related to the tile for which processing has been performed in a past and the intermediate data, and a processing unit that performs processing by the plurality of convolution layers and processing by the residual layer that adds the past layer data to the processing result.

公开(公告)号：US20240232593A9

公开(公告)日：2024-07-11

申请号：US18269528

申请日：2020-12-28

Applicant: NIPPON TELEGRAPH AND TELEPHONE CORPORATION

Inventor： Saki HATTA ,  Hiroyuki UZAWA ,  Shuhei YOSHIDA ,  Daisuke KOBAYASHI ,  Yuya OMORI ,  Ken NAKAMURA ,  Koyo NITTA

IPC: G06N3/063

CPC classification number: G06N3/063

Abstract: In a data processing device, a fixed-point position control unit determines, as first control. The fixed-point position control unit causes a detection calculation unit to perform calculation processing on processing target data at a processing point in time. The saturation rate control unit instructs, as second control to be repeated by the fixed-point position control unit, the fixed-point position control unit to move at least the fixed-point position as control to increase a lower limit saturation rate proportional to a magnitude of a counted lower limit counter value with respect to a result of the first control. The fixed-point position control unit performs, as the second control, a predetermined determination on the basis of the instruction from the saturation rate control unit and the metadata, determines the fixed-point position moved for each layer, and causes calculation processing to be performed.