DETAILED ACTION
[1] Remarks
I. The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
II. The amendment filed on 06/23/2026 has entered and made of record.
III. Claims 1-20 are pending and have been examined, where claims 1-17 is/are rejected, claim 18-20 is/are objected. Explanations will be provided below.
IV. Inventor and/or assignee search were performed and determined no double patenting rejection(s) is/are necessary.
V. Patent eligibility (updated in 2019) shown by the following: Claims 1-20 pass patent eligibility test because there is/are no limitation or a combination of limitations amounting to an abstract idea. Also, the following limitation or the combinations of the limitations:
apply an inverse Fast Fourier Transform to the matrix-vector product for each pixel in each image patch to convert each image patch to a spatial domain; and reconstruct, after application of the inverse Fast Fourier Transform, a convolved version of the input image by summing overlapping portions of each image patch together;
apply an inverse Fast Fourier Transform to the matrix-vector product for each pixel in each image patch to convert each image patch to a spatial domain;
effect a transformation or a reduction of a particular article to a different state or thing / adds a specific limitation(s) other than what is well-understood, routine and conventional in the field, or adding unconventional steps that confine the claim to a particular useful application and providing improvements to the technical field of deep learning accelerator which recite additional elements that integrate the judicial exception into a practical application and amounting significant more.
VI. There are no PCT associated with the current application.
[2] Response to Arguments
The arguments presented by the applicant have been considered and are found convincing.
An updated search was performed and all claims determined to be allowable. Details are shown below.
[3] Claim Interpretation
The following is a quotation of 35 U.S.C. 112(f):
(f) Element in Claim for a Combination. – An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The following is a quotation of pre-AIA 35 U.S.C. 112, sixth paragraph:
An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
Use of the word “means” (or “step for”) in a claim with functional language creates a rebuttable presumption that the claim element is to be treated in accordance with 35 U.S.C. 112(f) (pre-AIA 35 U.S.C. 112, sixth paragraph). The presumption that 35 U.S.C. 112(f) (pre-AIA 35 U.S.C. 112, sixth paragraph) is invoked is rebutted when the function is recited with sufficient structure, material, or acts within the claim itself to entirely perform the recited function. Absence of the word “means” (or “step for”) in a claim creates a rebuttable presumption that the claim element is not to be treated in accordance with 35 U.S.C. 112(f) (pre-AIA 35 U.S.C. 112, sixth paragraph). The presumption that 35 U.S.C. 112(f) (pre-AIA 35 U.S.C. 112, sixth paragraph) is not invoked is rebutted when the claim element recites function but fails to recite sufficiently definite structure, material or acts to perform that function.
Claim elements in this application that use the word “means” (or “step for”) are presumed to invoke 35 U.S.C. 112(f) except as otherwise indicated in an Office action. Similarly, claim elements that do not use the word “means” (or “step for”) are presumed not to invoke 35 U.S.C. 112(f) except as otherwise indicated in an Office action.
Claim(s) 1-17 are not interpreted under 35 U.S.C. 112(f) or pre-AIA U.S.C. 112 6th paragraph because of the following reason(s): limitations are modified by sufficient structure or material for performing the claimed function.
Claim(s) 18-20 do not require 35 U.S.C. 112(f) or pre-AIA U.S.C. 112 6th paragraph interpretation because they are method claims and / or they are CRM claims.
Upon examination of the specification and claims, the examiner has determined, under the best understanding of the scope of the claim(s), rejection(s) under 35 U.S.C. 112(a)/(b) is not necessitated because of the following reasons: sufficient support are provided in the written description / drawings of the invention.
[4] Allowable Subject Matter
Claims 1-17 are allowable / patentable. The following is an examiner’s statement of reasons for allowance by comparing claims to closest references. The references are divided into primary and secondary, where primary would have been utilized in a USC 102 or main USC 103 reference and secondary would had been utilized a secondary USC 103 reference, but these references do not cover enough of the claim’s scope to warrant a rejection.
Claims 1-20 are allowable / patentable. The following is an examiner’s statement of reasons for allowance by comparing claims to closest references. The references are divided into primary and secondary, where primary would have been utilized in a USC 102 or main USC 103 reference and secondary would had been utilized a secondary USC 103 reference, but these references do not cover enough of the claim’s scope to warrant a rejection.
Primary reference, Chitsaz et al. Acceleration of Convolutional Neural Network Using FFT-Based Split Convolutions, arXiv, 27 Mar 2020) discloses a system, comprising: a memory; and a deep learning accelerator configured to execute instructions from the memory, wherein the deep learning accelerator (see 4. Experimental Results first paragraph, FGPA is employed as the deep learning accelerator) is configured to;
divide an input image associated with an artificial intelligence task into equally-sized partially overlapping image patches (see figure 2 illustration below, partially overlapping, dotted lines);
apply, by utilizing a neural network, a Fast Fourier Transform to each of the equally-sized partially overlapping image patches (see figure 2, is a FFT-based processing of CNN using splitting); and
compute, for each pixel in each image patch of the equally-size partially overlapping image patches, a matrix-vector product between at least one channel of each image patch and a matrix from a corresponding pixel location in an image filter (see figure 1 illustration below, matrix dot product multiplication):
PNG
media_image1.png
555
1327
media_image1.png
Greyscale
.
Abtahi (T. Abtahi, C. Shea, A. Kulkarni and T. Mohsenin, "Accelerating Convolutional Neural Network with FFT on Embedded Hardware," in IEEE Transactions on Very Large Scale Integration (VLSI) Systems, vol. 26, no. 9, pp. 1737-1749, Sept. 2018) discloses a system, comprising:
a memory; and a deep learning accelerator configured to execute instructions from the memory (see A. PENC Many-core Overview and Key Features), wherein the deep learning accelerator is configured to;
PNG
media_image2.png
94
517
media_image2.png
Greyscale
divide an input image associated with an artificial intelligence task into equally-sized
apply, by utilizing a neural network, a Fast Fourier Transform to each of the equally-sized
compute, for each pixel in each image patch of the equally-size
apply an inverse Fast Fourier Transform to the matrix-vector product for each pixel in each image patch to convert each image patch to a spatial domain (see figure 3, 2D IFFT); and
reconstruct, after application of the inverse Fast Fourier Transform, a convolved version of the input image by multiplying
PNG
media_image3.png
446
879
media_image3.png
Greyscale
.
Primary reference, Srinivasan (US 7751482) discloses
divide an input image associated with an artificial intelligence task into equally-sized partially overlapping image patches (see figure 4 illustration below);
PNG
media_image4.png
419
786
media_image4.png
Greyscale
;
apply,
PNG
media_image5.png
124
420
media_image5.png
Greyscale
compute, for each pixel in each image patch of the equally-size partially overlapping image patches, a matrix-vector product between at least one channel of each image patch and a matrix from a corresponding pixel location in an image filter (see column 5, lines 55-63, where F1 is read as the channel of each image patch and F2 is read as a matrix from a corresponding pixel location):
PNG
media_image6.png
214
632
media_image6.png
Greyscale
.
apply an inverse Fast Fourier Transform to the matrix-vector product for each pixel in each image patch to convert each image patch to a spatial domain (see equation 6 illustration below):
PNG
media_image7.png
95
671
media_image7.png
Greyscale
.
Srinivasan is silent in disclosing reconstruct, after application of the inverse Fast Fourier Transform, a convolved version of the input image by summing overlapping portions of each image patch together (see equation 6 where there are no summations to obtain the output).
[5] Grounds of Rejection
Claim Rejections - 35 USC § 103
1. The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
2. Claims 18-19 is/are rejected under 35 U.S.C. 103 as being unpatentable over Chitsaz et al. Acceleration of Convolutional Neural Network Using FFT-Based Split Convolutions, arXiv, 27 Mar 2020 in view of Srinivasan (US 7751482).
Regarding claim 1, Chitsaz discloses a method, comprising:
splitting an input image into partially overlapping image patches (see figure 2 illustration below, partially overlapping, dotted lines);
applying, by utilizing a deep learning accelerator, a Fast Fourier Transform to each of the partially overlapping image patches and to an image filter (see figure 2, is a FFT-based processing of CNN using splitting);
computing, for each pixel in each image patch of partially overlapping image patches, a matrix-vector product between at least one channel of each image patch and a matrix from a corresponding pixel location in the image filter (see figure 1 illustration below, matrix dot product multiplication):
PNG
media_image1.png
555
1327
media_image1.png
Greyscale
.
Chitsaz is silent in disclosing apply an inverse Fast Fourier Transform to the matrix-vector product for each pixel in each image patch to convert each image patch to a spatial domain
Srinivasan discloses apply an inverse Fast Fourier Transform to the matrix-vector product for each pixel in each image patch to convert each image patch to a spatial domain (see equation 6 illustration below, δ(x0,y0) is in spatial coordinates):
PNG
media_image7.png
95
671
media_image7.png
Greyscale
.
It would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to include apply an inverse Fast Fourier Transform to the matrix-vector product for each pixel in each image patch to convert each image patch to a spatial domain because operations like blurring, sharpening, or feature extraction require spatial convolution, which is computationally expensive, but the complex frequency matrix operation simplifies into an efficient, element-wise multiplication which save computational time.
Regarding claim 19, Srinivasan discloses the method of claim 18, further comprising: reconstructing, after application of the inverse Fast Fourier Transform, a convolved version of the input image (see equation 6, where F-1 is read as the inverse FFT, where the F1(u,v) F2*(u,v) the convolve version of the input image); and applying a Fast Fourier Transform to the image filter to convert the filter to a frequency domain (equation 4, F2(u,v) is the frequency component of equation 3, f2(x,y)). In addition, multiplication in the frequency domain is equivalent to convolution in the spatial domain, making it a much faster way to process large images.
[6] Claim Objections
Claim(s) 20 is/are objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
With regards to claim 20, the examiner cannot find any applicable prior art providing teachings for the following limitation(s): the method of claim 18, further comprising reconstructing the convolved version of the input image by discarding overlapping regions of the image patches and combining retained portions of the image patches, or by summing the overlapping regions of the image patches together; in combination with the rest of the limitations of claim 18.
FUJIMURA (US 20160056883) discloses image components are generated by up-sampling during multiplexing. However, as in the case of the alias components, even if transition regions of the filter characteristics of the multiplexing units are expanded until immediately before the image components overlap a main signal component, the image components can be removed (see paragraph 141), but is silent in disclosing the method of claim 18, further comprising reconstructing the convolved version of the input image by discarding overlapping regions of the image patches and combining retained portions of the image patches, or by summing the overlapping regions of the image patches together.
CONTACT INFORMATION
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ALEX LIEW (duty station is located in New York City) whose telephone number is (571)272-8623 (FAX 571-273-8623), cell (917)763-1192 or email alexa.liew@uspto.gov. Please note the examiner cannot reply through email unless an internet communication authorization is provided by the applicant. The examiner can be reached anytime.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, MISTRY ONEAL R, can be reached on (313)446-4912. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/ALEX KOK S LIEW/Primary Examiner, Art Unit 2674 Telephone: 571-272-8623
Date: 9/3/26
Prosecution Insights