Information processing device, vector data processing method, and recording medium

• US 10,459,902 B2
• Filed: 06/11/2015
• Issued: 10/29/2019
• Est. Priority Date: 06/19/2014
• Status: Active Grant

• Current Assignee: NEC Corporation
• Original Assignee: NEC Corporation
• Inventors: Ishizaka, Kazuhisa
• Primary Examiner(s): Lewis, Cheryl
• Application Number:
  US15/315,469
• Publication Number:
  US 20170199907A1
• Time in Patent Office: 1,601 Days
• CPC Class Codes: G06F 16/2237 Vectors, bitmaps or matrices
• CPC Class Codes: G06F 17/16 Matrix or vector computation , e.g. matrix-matrix or matrix-vector multiplication, matrix factorization matrix transposition G06F7/78
• CPC Class Codes: G06F 18/22 Matching criteria, e.g. proximity measures
• CPC Class Codes: G06F 7/02 Comparing digital values G06F7/06, G06F7/22, G06F7/38 take precedence
• CPC Class Codes: G06T 7/00 Image analysis
• CPC Class Codes: G06T 9/008 Vector quantisation

Abstract

Matching processing between pieces of vector data is accelerated. A matching device 100 performs, for a plurality of pieces of vector data each having a plurality of dimensions, a predetermined operation pertaining to each dimension of each piece of vector data. The matching device 100 includes a collective operation unit 150 and an individual operation unit 160. The collective operation unit 150 performs the predetermined operation pertaining to a specific dimension among the plurality of dimensions by a vector operation for different pieces of vector data in the plurality of pieces of vector data. The individual operation unit 160 performs the predetermined operation pertaining to each dimension other than the specific dimension for a piece of vector data that satisfies a predetermined condition among the plurality of pieces of vector data.

Claims

1. An information processing device that performs, for a plurality of pieces of vector data each having a plurality of dimensions, a predetermined operation pertaining to each dimension of each piece of vector data, the information processing device comprising:

a collective operation unit that performs the predetermined operation pertaining to a specific dimension among the plurality of dimensions by a vector operation for different pieces of vector data in the plurality of pieces of vector data; and

an individual operation unit that performs the predetermined operation pertaining to each dimension other than the specific dimension for a piece of vector data that satisfies a predetermined condition among the plurality of pieces of vector data.

2. The information processing device according to claim 1, wherein

the specific dimension is a dimension, of which order of performing the predetermined operation is equal to or less than a predetermined value, among the plurality of dimensions.

3. The information processing device according to claim 1, further comprising:

a dimension determination unit that determines the specific dimension depending on a performing status of the predetermined operation in at least one of the collective operation unit and the individual operation unit.

4. The information processing device according to claim 3, wherein, the dimension determination unit includes, when the number of times or proportion of the predetermined operation performed by the individual operation unit is equal to or more than a predetermined value for a dimension other than the specific dimension, the dimension in the specific dimension.

5. The information processing device according to claim 1, wherein

the predetermined operation includes a calculation of a difference between pieces of vector data for each dimension and a calculation of an accumulated value of the difference, and

the individual operation unit performs the predetermined operation for vector data of which the accumulated value of the difference is less than a predetermined truncation threshold as the predetermined condition.

6. A vector data processing method for performing, for a plurality of pieces of vector data each having a plurality of dimensions, a predetermined operation pertaining to each dimension of each piece of vector data, the vector data processing method comprising:

performing collective operation processing that performs the predetermined operation pertaining to a specific dimension among the plurality of dimensions by a vector operation for different pieces of vector data in the plurality of pieces of vector data; and

performing individual operation processing that performs the predetermined operation pertaining to each dimension other than the specific dimension for a piece of vector data that satisfies a predetermined condition among the plurality of pieces of vector data.

7. The vector data processing method according to claim 6, wherein

the specific dimension is a dimension, of which order of performing the predetermined operation is equal to or less than a predetermined value, among the plurality of dimensions.

8. The vector data processing method according to claim 6, further comprising:

determining the specific dimension depending on a performing status of the predetermined operation in at least one of the collective operation processing and the individual operation processing.

9. The vector data processing method according to claim 8, wherein, when the number of times or proportion of the predetermined operation performed by the individual operation processing is equal to or more than a predetermined value for a dimension other than the specific dimension, the dimension is included in the specific dimension.

10. A non-transitory computer readable storage medium recording thereon a program for a computer for performing, for a plurality of pieces of vector data each having a plurality of dimensions, a predetermined operation pertaining to each dimension of each piece of vector data, the program causing the computer to execute processes comprising:

a process for performing the predetermined operation pertaining to a specific dimension among the plurality of dimensions by a vector operation for different pieces of vector data in the plurality of pieces of vector data; and

a process for performing the predetermined operation pertaining to each dimension other than the specific dimension for a piece of vector data that satisfies a predetermined condition among the plurality of pieces of vector data.