Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Claims 1-15 are presented for examination.
Specification
The specification is objected to as failing to provide proper antecedent basis for the claimed subject matter. See 37 CFR 1.75(d)(1) and MPEP § 608.01(o). Correction of the following is required:
The use of the terms Intel (Paragraph 44), Nvidia (Paragraph 44), and Pytorch (Paragraph 44) are trade names or marks used in commerce, have been noted in this application. The terms should be accompanied by the generic terminology; furthermore the term should be capitalized wherever it appears or, where appropriate, include a proper symbol indicating use in commerce such as ™, SM , or ® following the term.
Although the use of trade names and marks used in commerce (i.e., trademarks, service marks, certification marks, and collective marks) are permissible in patent applications, the proprietary nature of the marks should be respected and every effort made to prevent their use in any manner which might adversely affect their validity as commercial marks.
The disclosure is objected to because of the following informalities:
The phrase "stitching the depth features along a channel domain" recited in paragraphs 7, 10, 25, and 33 should be changed to where "stitching" should read "concatenating," and "channel domain" should read "channel dimension" for consistency.
Paragraphs 8 and 22 recite “full connection layers,” which should be changed to "fully-connected layers."
In paragraph 24, the displayed loss function recites "when = 1". The variable "y" appears to be omitted ("when y = 1").
Appropriate correction is required.
Drawings
The drawings are objected to because the drawings (FIGS. 1–2) are rendered in greyscale with shaded/photographic fill and Fig. 3 contains text in a font size that is too small to be legible. Corrected drawing sheets in compliance with 37 CFR 1.121(d) are required in reply to the Office action to avoid abandonment of the application. Any amended replacement drawing sheet should include all of the figures appearing on the immediate prior version of the sheet, even if only one figure is being amended. The figure or figure number of an amended drawing should not be labeled as “amended.” If a drawing figure is to be canceled, the appropriate figure must be removed from the replacement sheet, and where necessary, the remaining figures must be renumbered and appropriate changes made to the brief description of the several views of the drawings for consistency. Additional replacement sheets may be necessary to show the renumbering of the remaining figures. Each drawing sheet submitted after the filing date of an application must be labeled in the top margin as either “Replacement Sheet” or “New Sheet” pursuant to 37 CFR 1.121(d). If the changes are not accepted by the examiner, the applicant will be notified and informed of any required corrective action in the next Office action. The objection to the drawings will not be held in abeyance.
Claim Objections
Claims 1-15 objected to because of the following informalities:
Claim 1 (steps 1 and 7) recites “generating a certain number of positive samples and a certain number of negative samples according to target location information provided in an initial frame… generating a certain number of positive samples and a certain number of negative samples using the current frame according to the credit score S…”. It is unclear whether the certain number of positive and negative samples refer to the same number or not. Examiner suggests that the two instances of “certain” in the first limitation be changed to “first” and that the two instances of “certain” in the second limitation be changed to “second”.
Claim 2 recites “full connection layers,” which should be changed to “fully-connected layers.”
Claim 4 recites "...wherein C is a dimension of the feature map**.** normalizing the similarity matrix Λ by a function softmax...". The claim contains a period followed by the lower-case sentence fragment "normalizing the similarity matrix...". The internal period should be replaced with appropriate punctuation (e.g., a semicolon) and the limitations recited as a continuous single sentence.
Claim 4 recites "(·)^T is an matrix transpose operation, and concat(·,·) represents a series operation." The phrase "an matrix" should read "a matrix".
Claim 1 recited GoogleNet which should be changed to recite GoogLeNet instead.
Appropriate correction is required.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 1-15 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
The limitation “…a scale 11 factor set to prevent a corresponding value from being too large…” in claim 4 is a relative term which renders the claim indefinite. The term “too large” is not defined by the claim, the specification does not provide a standard for ascertaining the requisite degree, and one of ordinary skill in the art would not be reasonably apprised of the scope of the invention. The term "too large" is a term of degree with no standard, and the specification does not provide any definition for ascertaining the degree of being too large.
Claim 1 (steps 6 and 8) recites “…to obtain a credit score S of the prediction result of the current frame… the credit score of the prediction result of each historical frame…”. It is unclear whether the credit score S is obtained using the current frame or the historical frame. Similarly, claim 5 recites “the prediction results of all historical frames,” which should also address this clarity issue.
Claim 4 recites the limitation “calculating a similarity between each pixel of the memory feature f" and the current frame feature f C to obtain a similarity matrix A” in lines 2-3. There is insufficient antecedent basis for this limitation in the claim. Claim 1 introduces "memory features" (step 2) and "current frame features" (step 3) in the plural. It is unclear whether the singular feature is the same as the previously recited plural features.
Claim 4 recites the limitation "a scale 11 factor set to prevent a corresponding value from being too large wherein C is a dimension of the feature map." in lines 8-9. There is insufficient antecedent basis for this limitation in the claim. It is unclear which feature map (e.g., location information feature map, memory feature map, current frame feature map, etc.) is referred to.
Claim 5 recites the variables n and t in the equation, which are not defined in the claim. The variable n is also not recited in the specification as well. There is insufficient antecedent basis for these variables in the claim.
Claims 11-15 each recite the limitation "when being executed by the processor, the computer program implements " in lines 1-2. There is insufficient antecedent basis for this limitation in the claim.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 11-15 are rejected under 35 U.S.C. 101 because claims 11-15 recite the phrase “computer readable medium”. “When the BRI encompasses transitory forms of signal transmission, a rejection under 35 U.S.C. 101 as failing to claim statutory subject matter would be appropriate. Thus, a claim to a computer readable medium that can be a compact disc or a carrier wave covers a non-statutory embodiment and therefore should be rejected under 35 U.S.C. 101 as being directed to non-statutory subject matter. See, e.g., Mentor Graphics v. EVE-USA, Inc., 851 F.3d at 1294-95, 112 USPQ2d at 1134 (claims to a “machine-readable medium” were non-statutory, because their scope encompassed both statutory random-access memory and non-statutory carrier waves).” See MPEP 2106.03(II).
Claims 1-15 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an
abstract idea without significantly more.
Claim 1
Step 1: The claim recites a method; therefore, it is directed to the statutory category of a process.
Step2A Prong 1: The claim recites, inter alia:
step 1, generating a certain number of positive samples and a certain number of negative samples according to target location information provided in an initial frame: This limitation encompasses a mental process of evaluation/judgement/opinion to generate positive and negative samples based on target location information in a frame.
initializing the credit allocation network by a guiding focus loss function: This limitation recites a mathematical concept because it involves initializing the credit allocation network through positive and negative samples with the process of using a focus loss function. Paragraphs 22 and 23 of the instant specification describe this process where “[t]he credit allocation network… are initialized through 500 positive samples and 2000 negative samples…The guiding focus loss function is expressed as:
PNG
media_image1.png
131
495
media_image1.png
Greyscale
”.
and stitching the depth features along a channel domain to obtain memory features: This limitation recites a mathematical concept because it involves concatenating the extracted depth features with the channel dimension/domain. See Paragraphs 33-34 which involves “normalizing the similarity matrix A by a function soft maX , and multiplying the normalized similarity matrix by the memory features to adaptively read the target information… and stitching the current frame feature and a read-out feature along the channel domain to obtain the location information feature map:
PNG
media_image2.png
60
482
media_image2.png
Greyscale
.”
step 3, reading a next frame image as the current frame to be tracked, cutting a current frame image according to the target location information in a previous frame image: This limitation encompasses a mental process of evaluation/judgement/opinion of determining a next frame image, cropping the image frame based on target location information from a previous image frame.
step 7, generating a certain number of positive samples and a certain number of negative samples using the current frame according to the credit score S: This limitation encompasses a mental process of evaluation/judgement/opinion to generate positive and negative samples based on the credit score of the current image frame.
Step2A Prong 2: This judicial exception is not integrated into a practical application because the
additional elements are as follows:
step 2, extracting depth features of all memory samples in the memory pool using a pre-trained GoogleNet network: Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea (MPEP 2106.05(f)).
and inputting the cut current frame image to the GoogleNet network to obtain current frame features: Mere data gathering recited at a high level of generality, and thus is an insignificant extra-solution activity (MPEP 2106.05(g)).
step 4, inputting the memory features and the current frame features into a time-space memory network: Mere data gathering recited at a high level of generality, and thus is an insignificant extra-solution activity (MPEP 2106.05(g)).
step 5, reading the location information feature map of the current frame using a single convolution network to generate a classification, a centrality and a regression response map to predict the target location information in the current frame: Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea (MPEP 2106.05(f)).
step 6, inputting the current frame image and the predicted target location information into the credit allocation network…: Mere data gathering recited at a high level of generality, and thus is an insignificant extra-solution activity (MPEP 2106.05(g)).
…the credit allocation network to obtain a credit score S of the prediction result of the current frame: Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea (MPEP 2106.05(f)).
and updating the credit allocation network online by the guiding focus loss function: Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea (MPEP 2106.05(f)).
and putting the initial frame into a memory pool as a first target memory sample… step 8, updating the memory samples in the memory pool according to the credit score of the prediction result of each historical frame: This limitation is merely a post-solution step of storing the data—a nominal addition to the claim that does not meaningfully limit the claim. The method storing is recited at a high level of generality. Simply implementing the abstract idea in a generic method is not a practical application of the abstract idea. Therefore, storing step is an insignificant extra-solution activity. See MPEP 2106.05(g).
and step 9, circularly performing step 2 to step 7 until a video sequence is traversed to complete the target tracking: The limitation amounts to merely indicating a field of use or technological environment in which to apply a judicial exception. This does not amount to significantly more than the exception itself (MPEP 2106.05(h)).
and querying the target location information in the current frame using a memory frame to obtain a location information feature map: Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea a (MPEP 2106.05(f)).
Step 2B: The claim does not include additional elements that are sufficient to amount to significantly
more than the judicial exception because the additional elements are as follows:
step 2, extracting depth features of all memory samples in the memory pool using a pre-trained GoogleNet network: Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea and cannot provide inventive concept (MPEP 2106.05(f)).
and inputting the cut current frame image to the GoogleNet network to obtain current frame features: The additional element of “inputting” does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea. As discussed above with respect to integration of the abstract idea into a practical application, the additional element of receiving steps amounts to no more than mere data gathering. This element amounts to receiving data over a network and is well-understood, routine, conventional activity. See MPEP 2106.05(d), subsection II (i). This cannot provide an inventive concept.
step 4, inputting the memory features and the current frame features into a time-space memory network: The additional element of “inputting” does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea. As discussed above with respect to integration of the abstract idea into a practical application, the additional element of receiving steps amounts to no more than mere data gathering. This element amounts to receiving data over a network and is well-understood, routine, conventional activity. See MPEP 2106.05(d), subsection II (i). This cannot provide an inventive concept.
step 5, reading the location information feature map of the current frame using a single convolution network to generate a classification, a centrality and a regression response map to predict the target location information in the current frame: Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea and cannot provide inventive concept (MPEP 2106.05(f)).
step 6, inputting the current frame image and the predicted target location information into the credit allocation network…: The additional element of “inputting” does not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea. As discussed above with respect to integration of the abstract idea into a practical application, the additional element of receiving steps amounts to no more than mere data gathering. This element amounts to receiving data over a network and is well-understood, routine, conventional activity. See MPEP 2106.05(d), subsection II (i). This cannot provide an inventive concept.
…the credit allocation network to obtain a credit score S of the prediction result of the current frame: Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea and cannot provide inventive concept (MPEP 2106.05(f)).
and updating the credit allocation network online by the guiding focus loss function: Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea and cannot provide inventive concept (MPEP 2106.05(f)).
and putting the initial frame into a memory pool as a first target memory sample… step 8, updating the memory samples in the memory pool according to the credit score of the prediction result of each historical frame: This limitation amounts to storing information in memory, Versata Dev. Group, Inc. v. SAP Am., Inc., 793 F.3d 1306, 1334, 115 USPQ2d 1681, 1701 (Fed. Cir. 2015); OIP Techs., 788 F.3d at 1363, 115 USPQ2d at 1092-93; See MPEP 2106.05(d) (II)(iv). The courts have recognized the computer functions of storing as well‐understood, routine, and conventional function when they are claimed in a merely generic manner (e.g., at a high level of generality) or as insignificant extra-solution activity.
and step 9, circularly performing step 2 to step 7 until a video sequence is traversed to complete the target tracking: The limitation amounts to merely indicating a field of use or technological environment in which to apply a judicial exception. This does not amount to significantly more than the exception itself which cannot provide inventive concept (MPEP 2106.05(h)).
and querying the target location information in the current frame using a memory frame to obtain a location information feature map: Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea and cannot provide inventive concept (MPEP 2106.05(f)).
The elements in combination as an ordered whole still do not amount to significantly more than the judicial exception (i.e., the abstract ideas of mental processes for generating positive and negative samples and selecting and updating memory samples, and mathematical concepts including the guiding focus loss function, the similarity-matrix and softmax read-out, the channel-wise concatenation, and the memory-sampling formula for single-target tracking). The claim merely describes standard data processing steps (reading and cropping image frames, inputting features into networks, storing and updating memory samples in a memory pool, and outputting a credit score). The recitation of a pre-trained GoogleNet network, a time-space memory network, a single convolution network, a credit allocation network, and a generic processor and memory merely indicates a technological environment in which the abstract ideas are applied, without improving the functioning of a computer or the target-tracking technology itself.
Therefore, the claim as a whole remains focused on the abstract idea and fails Step 2B of the eligibility
analysis.
Claim 2
Step 1: A process, as above.
Step2A Prong 1: This claim does not recite an additional abstract idea, but the claim depends on
claim 1, which recites an abstract idea.
Step2A Prong 2: This judicial exception is not integrated into a practical application because the
additional elements are as follows:
the credit allocation network comprises three convolution layers, two full connection layers and a secondary classification layer: The limitation amounts to merely indicating a field of use or technological environment in which to apply a judicial exception. This does not amount to significantly more than the exception itself (MPEP 2106.05(h)).
the convolution layers are fixed by offline-training parameters, and the full connection layers and the secondary classification layer are initialized by 500 positive samples and 2000 negative samples: The limitation amounts to merely indicating a field of use or technological environment in which to apply a judicial exception. This does not amount to significantly more than the exception itself (MPEP 2106.05(h)).
Step 2B: The claim does not include additional elements that are sufficient to amount to significantly
more than the judicial exception because the additional elements are as follows:
the credit allocation network comprises three convolution layers, two full connection layers and a secondary classification layer: The limitation amounts to merely indicating a field of use or technological environment in which to apply a judicial exception. This does not amount to significantly more than the exception itself which cannot provide inventive concept (MPEP 2106.05(h)).
the convolution layers are fixed by offline-training parameters, and the full connection layers and the secondary classification layer are initialized by 500 positive samples and 2000 negative samples: The limitation amounts to merely indicating a field of use or technological environment in which to apply a judicial exception. This does not amount to significantly more than the exception itself which cannot provide inventive concept (MPEP 2106.05(h)).
Even when considered in combination, these additional elements represent mere instructions to apply
an exception and therefore do not provide an inventive concept. The claim is ineligible.
Claim 3
Step 1: A process, as above.
Step2A Prong 1: The claim recites, inter alia:
the guiding focus loss function is expressed as:
PNG
media_image3.png
93
399
media_image3.png
Greyscale
wherein P is a predicted output, y =1 represents a positive sample, t is the number of iterations, and A is an initial focusing factor: This limitation recites a mathematical concept describing the equation for the guiding focus loss function.
Step 2A Prong Two and Step 2B: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception under step 2B. Thus, the judicial exception is not integrated into a practical application (see MPEP 2106.04(d) I.), failing step 2A prong 2. The claim is ineligible.
Even when considered in combination, these additional elements represent mere instructions to apply
an exception and therefore do not provide an inventive concept. The claim is ineligible.
Claim 4
Step 1: A process, as above.
Step2A Prong 1: The claim recites, inter alia:
wherein in step 4, the process of obtaining the location information feature map comprises: calculating a similarity between each pixel of the memory feature f’’ and the current frame feature f C to obtain a similarity matrix A, wherein each element of the similarity matrix A is calculated according to:
PNG
media_image4.png
96
342
media_image4.png
Greyscale
wherein, i is an index of each pixel on a memory feature map, j is an index of each pixel on the current frame feature map, a represents a dot product of vectors, and v/C is a scale 11 factor set to prevent a corresponding value from being too large wherein C is a dimension of the feature map: This limitation recites a mathematical concept calculating a similarity between each pixel of the memory feature of the memory feature and the current frame feature.
normalizing the similarity matrix A by a function softmax, and multiplying the normalized similarity matrix by the memory features to adaptively read the target information stored therein, and stitching the current frame feature and a read-out feature along the channel domain to obtain the location information feature map
PNG
media_image5.png
44
382
media_image5.png
Greyscale
wherein (') an matrix transpose operation, and CONCat(-,') represents a series operation along a channel dimension: This limitation recites a mathematical concept since the equation concatenates depth features that applies the softmax to the similarity matrix A.
Step 2A Prong Two and Step 2B: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception under step 2B. Thus, the judicial exception is not integrated into a practical application (see MPEP 2106.04(d) I.), failing step 2A prong 2. The claim is ineligible.
Even when considered in combination, these additional elements represent mere instructions to apply
an exception and therefore do not provide an inventive concept. The claim is ineligible.
Claim 5
Step 1: A process, as above.
Step2A Prong 1: The claim recites, inter alia:
the credit scores of the prediction results of all historical frames are divided into five ranges, and the frame with the highest credit score and greater than 0 in each range is selected as a memory sample to be put into the memory pool: This limitation encompasses a mental process of evaluation/judgement/opinion to divide historical frames into ranges, and select a memory sample for each range that satisfies a threshold.
and a sampling method of the memory samples is described as
PNG
media_image6.png
83
598
media_image6.png
Greyscale
wherein <p_i is a sequence index of the memory sample in the historical frame, and i ∈ {1, 2, 3, 4, 5}: This limitation recites a mathematical concept describing how the sampling method is calculated.
Step 2A Prong Two and Step 2B: The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception under step 2B. Thus, the judicial exception is not integrated into a practical application (see MPEP 2106.04(d) I.), failing step 2A prong 2. The claim is ineligible.
Even when considered in combination, these additional elements represent mere instructions to apply
an exception and therefore do not provide an inventive concept. The claim is ineligible.
Claim 6
Step 1: The claim recites an electronic apparatus; therefore, it is directed to the statutory category of machine.
Step2A Prong 1: The claim implements the steps of the single-target tracking method of claim 1, which recites abstract ideas.
Step2A Prong 2: This judicial exception is not integrated into a practical application because the
additional elements are as follows:
[a]n electronic apparatus having a memory, a processor, and a computer program stored in the memory and capable of running in the processor, wherein, when being executed by the processor: Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea (MPEP 2106.05(f)).
Step 2B: The claim does not include additional elements that are sufficient to amount to significantly
more than the judicial exception because the additional elements are as follows:
[a]n electronic apparatus having a memory, a processor, and a computer program stored in the memory and capable of running in the processor, wherein, when being executed by the processor: Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea and cannot provide inventive concept (MPEP 2106.05(f)).
Even when considered in combination, these additional elements represent mere instructions to apply
an exception and therefore do not provide an inventive concept. The claim is ineligible.
Claim 7 recites identical limitations to claim 2. Therefore, claim 7 is rejected using the same rationale as claim 2.
Claim 8 recites identical limitations to claim 3. Therefore, claim 8 is rejected using the same rationale as claim 3.
Claim 9 recites identical limitations to claim 4. Therefore, claim 9 is rejected using the same rationale as claim 4.
Claim 10 recites identical limitations to claim 5. Therefore, claim 10 is rejected using the same rationale as claim 5.
Claim 11
Step 1: The claim recites a computer-readable storage medium; therefore, it is directed to the statutory category of article of manufacture.
Step2A Prong 1: The claim implements the steps of the single-target tracking method of claim 1, which recites abstract ideas.
Step2A Prong 2: This judicial exception is not integrated into a practical application because the
additional elements are as follows:
[a] computer-readable storage medium with a computer program stored thereon, wherein, when being executed by the processor, the computer program: Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea and cannot provide inventive concept (MPEP 2106.05(f)).
Step 2B: The claim does not include additional elements that are sufficient to amount to significantly
more than the judicial exception because the additional elements are as follows:
[a] computer-readable storage medium with a computer program stored thereon, wherein, when being executed by the processor, the computer program: Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea (MPEP 2106.05(f)).
Even when considered in combination, these additional elements represent mere instructions to apply
an exception and therefore do not provide an inventive concept. The claim is ineligible.
Claim 12 recites identical limitations to claim 2. Therefore, claim 12 is rejected using the same rationale as claim 2.
Claim 13 recites identical limitations to claim 3. Therefore, claim 13 is rejected using the same rationale as claim 3.
Claim 14 recites identical limitations to claim 4. Therefore, claim 14 is rejected using the same rationale as claim 4.
Claim 15 recites identical limitations to claim 5. Therefore, claim 15 is rejected using the same rationale as claim 5.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1-2, 4, 6-7, 9, 11-12, and 14 are rejected under 35 U.S.C. 103 as being unpatentable over Fu (“STMTrack: Template-free Visual Tracking with Space-time Memory Networks”, 2021) in view of Zhang (CN 115272401 A) and in view of Liu (“Learning Quality-aware Dynamic Memory for Video Object Segmentation”, 2022), and in further view of Hines (US 20220075782 A1).
Regarding claim 1,
Fu teaches [a] single-target tracking method based on a … network, comprising: (Page 1 Introduction of Fu, “The goal of visual tracking is to locate an object in the subsequent frames of a video given its initial annotation in the first frame… We propose a novel end-to-end memory-based tracking framework…”, Page 3 Section 3.1, “As shown in Fig. 2, the framework can be divided into three parts, a feature extraction network, a space-time memory network, and a head network.”
Fu teaches a single-target visual tracking method that locates an annotated target across the frames of a video.).
step 1, … and putting the initial frame into a memory pool as a first target memory sample (Page 5 Section 3.5, "the target from the first frame provides the most reliable information, while the tracked target from the previous frame has the most simi lar appearance to the target in the current frame. Therefore, for the current frame Ft, the memory frames hold the first frame F1, the previous frame Ft−1 and other N −2 frames…"
Fu stores the first annotated frame among the memory frames used to localize the target. The set of memory frames corresponds to the claimed memory pool and Fu’s retention of the first frame as the most reliable memory frame corresponds to putting the initial frame into the memory pool as a first target memory sample.).
step 2, extracting depth features of all memory samples in the memory pool using a pre-trained GoogleNet network, and stitching the depth features along a channel domain to obtain memory features (Page 5 Section 4.2, "We adopt GoogLeNet [43] as our backbone ϕm and ϕq.", §3.2, "input the sum to the latter layers of ϕm to generate T memory feature maps"; Fig. 2 and its Caption,
PNG
media_image7.png
392
394
media_image7.png
Greyscale
“The left part is the feature extraction network that consists of a memory branch (displayed in light green) and a query branch (displayed in light blue). The memory branch takes both memory frames and corresponding foreground-background label maps as inputs. “concat.” denotes the concatenation operation along the temporal dimension. "
Fu extracts deep features from each of the T memory frames using a GoogLeNet backbone and concatenates them across the memory frames to form the memory feature representation f^m. The GoogLeNet backbone corresponds to the pre-trained GoogleNet network, and Fu’s concatenation of the per-frame memory features corresponds to stitching the depth features to obtain memory features.)
step 3, reading a next frame image as the current frame to be tracked, cutting a current frame image according to the target location information in a previous frame image, and inputting the cut current frame image to the GoogleNet network to obtain current frame features (Page 3 Section 3.1, "the query frame is the current frame in a tracking sequence", Page 5 Section 4.1, "taking the target center as the center, we crop a square image patch with side length S from the original frame and resize the cropped image patch", Page 4 Section 3.2, "the query branch takes a query frame q as input and produces a feature map ϕq(q)"
Fu reads the current frame as the query frame, crops a search patch centered on the prior target location, and feeds the cropped patch through the GoogLeNet query branch to obtain query features. The query frame corresponds to the current frame to be tracked, the cropped square image patch corresponds to the cut current frame image, and the resulting query feature map ϕq(q) corresponds to the current frame features.).
step 4, inputting the memory features and the current frame features into a time-space memory network, and querying the target location information in the current frame using a memory frame to obtain a location information feature map (Page 3 Fig. 2 and its Caption,
PNG
media_image8.png
396
808
media_image8.png
Greyscale
“The middle part is the space-time memory network that retrieves the target information from multiple memory frames for the target localization in the query frame.”, Page 3 Section 3.1, "After the feature extraction, the space time memory network retrieves information related to the target from features of all memory frames, generating a synthetic feature map…", Page 4 Section 3.3, "…we concatenate the readout information and the query feature map f^q along the channel dimension to generate the final synthetic feature map y."
Fu feeds the memory features f^m and query features f^q into its space-time memory network, which reads target information out of the memory frames and produces a synthetic feature map y for localizing the target in the query frame. The synthetic feature map y corresponds to the location information feature map.)
step 5, reading the location information feature map of the current frame using a single convolution network to generate a classification, a centrality and a regression response map to predict the target location information in the current frame (Page 4-5 Section 3.4, "we design an anchor-free head network that contains a classification branch … and an anchor-free regression branch to directly estimate the target bounding box… producing the final classification response map Rcls… a sub-branch is forked after ωcls to generate a center-ness response map Rctr… In the regression branch, we pass y to another lightweight regression convolutional network ωreg and then reduce the dimensionality of the outputted features to 4 to generate a regression response map… for the target bounding box estimation.", See Page 3 Fig. 2,
PNG
media_image9.png
489
263
media_image9.png
Greyscale
Fu’s head network reads the synthetic feature map and produces a classification response map Rcls, a center-ness response map Rctr, and a regression response map Rreg to estimate the target box.).
step 9, circularly performing step 2 to step 7 until a video sequence is traversed to complete the target tracking (Page 1 Introduction, "The goal of visual tracking is to locate an object in the subsequent frames of a video", Page 5 Section 3.5, "for the current frame Ft, we select N memory frames from all historical frames (i.e. frame F1 to frame Ft−1) as memory frames for rich appearance in formation and strong generalization ability… For each frame in the whole tracking process, after obtaining Rcls, Rctr, and Rreg, the postprocessing is the same as [56]"
Fu repeats its feature-extraction, memory-read, and head-prediction operations for each frame in turn until the sequence is exhausted.)
Fu does not teach step 1, generating a certain number of positive samples and a certain number of negative samples according to target location information provided in an initial frame, initializing the credit allocation network by a guiding focus loss function… step 6, inputting the current frame image and the predicted target location information into the credit allocation network to obtain a credit score S of the prediction result of the current frame… step 7, generating a certain number of positive samples and a certain number of negative samples using the current frame according to the credit score S, and updating the credit allocation network online by the guiding focus loss function... step 8, updating the memory samples in the memory pool according to the credit score of the prediction result of each historical frame.
Zhang, in the same field of endeavor, teaches step 1, generating a certain number of positive samples and a certain number of negative samples according to target location information provided in an initial frame, initializing the … network by a guiding focus loss function… (Page 3 S1 of Zhang, "aiming at the first frame image in the video sequence, given the target coordinate information of the first frame image and intercepting the target template… collecting positive and negative samples according to the overlapping rate of the target coordinate, the focus loss metric is used for identifying the initialization of the enhanced memory model;", Page 3 S1.1, "…using the gradient descent method to train the three layers of network until convergence, wherein the loss function is focus loss measurement;"
Zhang collects positive and negative samples from the first frame using the given target coordinates and uses a focus loss measurement to initialize its identifying enhanced memory model. Zhang’s identifying enhanced memory model is a separately trained network that scores the reliability of the tracking result. The focus loss initialization corresponds to initializing that network by a guiding focus loss function.).
step 6, inputting the current frame image and the predicted target location information into the … network to obtain a … score S of the prediction result of the current frame (Page 3 S5, "inputting the coordinate information of the predicted target and the pixel searching area to judge the enhanced memory model to obtain the evaluation value Fmax (*), comparing the evaluation value with the set threshold value β…"
Zhang feeds the predicted target coordinates together with the pixel search area of the current frame into its identifying enhanced memory model to obtain an evaluation value Fmax for the prediction. The evaluation value Fmax () of the prediction result corresponds to the score S of the prediction result.).
step 7, generating a certain number of positive samples and a certain number of negative samples using the current frame according to the … score S, and updating the … network online by the guiding focus loss function (Page 3 S6, "using n-30 to n-1 frame collected positive and negative samples for re-online training and distinguishing enhanced memory model, again inputting the prediction target coordinate information and the pixel searching area input judging enhancement memory model to obtain the evaluation value Fmax (*)", Page 5 S1, "collecting positive and negative samples according to the overlapping rate of the target coordinate, initializing the distinguishing enhanced memory model by focus loss measurement, wherein the positive sample is 500, negative sample is 2000. In the subsequent tracking, each time collecting new positive and negative samples according to the coordinate information of the predicted target, wherein the positive sample is 50, negative sample 200;", Page 5 S1.1, "…fixing the front three layers of convolutional layer parameters, using the gradient descent method to train the three layers of network until convergence, wherein the loss function is focus loss measurement"
Zhang collects new positive and negative samples from the current/recent frames based on the predicted target and re-trains its evaluation model online using the focus loss measurement when the evaluation value falls below the threshold.).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date to combine Fu’s space-time memory tracker with Zhang’s reliability-scoring evaluation network that is initialized and updated by a focus loss in order to provide a more reliable way of tracking results for the target frame (S5–S7 of Zhang).
Fu in view of Zhang does not teach step 8, updating the memory samples in the memory pool according to the credit score of the prediction result of each historical frame.
Liu, in the same field of endeavor, teaches step 8, updating the memory samples in the memory pool according to the … score of the prediction result of each historical frame (Page 5 Sections 3.1 and 3.2, "the Quality Assessment Module (QAM) evaluates the quality of the segmentation result and decides whether the query frame can become a memory frame… To alleviate this problem and ensure the accuracy of the memory bank…we propose the Quality Assessment Module (QAM) to evaluate the segmentation quality and decide whether a frame can be added to the memory bank as a reference…", Page 7 Section 3.3, “it is necessary to limit the size of the memory bank and update it dynamically to adapt to new scenarios… we suggest dynamically updating the memory bank in accordance with these two principles (Algorithm.1). Specifically, when the memory bank reaches a certain storage limit, we will dynamically update the memory bank to handle different video scenes. For quantifying the temporal consistency and measuring the distance between each memory frame and the current frame, we compute the temporal consistency score SC as:
PNG
media_image10.png
59
208
media_image10.png
Greyscale
"
Liu computes a per-frame quality score and uses that score to decide which frames are admitted to and retained in the memory bank, dynamically updating the bank to exclude low-quality frames. Liu’s quality score corresponds to the claimed credit score of a historical frame's prediction result, and its score-gated admission/removal of memory frames corresponds to updating the memory samples in the memory pool according to that score.)
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date to combine Fu in view of Zhang’s memory-based tracker with Liu’s quality-score-gated memory admission in order to selectively store accurately tracked frames and prevent the error accumulation caused by indiscriminate memory updating (Introduction and Section 3.3 of Liu).
Fu in view of Zhang in further view of Liu does not teach [a] … method based on a credit allocation network … to obtain a credit score S of the prediction result …
Hines, in the same field of endeavor, teaches [a] … method based on a credit allocation network … to obtain a credit score S of the prediction result … (Paragraph 339, “Using a transformer neural network to predict a schedule may include the timestamps associated with scalar values (e.g., timestamps indicating when a credit utilization changed, timestamps indicating when an amount of a resource was consumed or allocated, etc.) into a corresponding set of time embeddings.”, Paragraph 344, “The first neural network model may be trained to increase the likelihood that the partial assignments will produce a greater objective value, where the objective value may correspond with a target value such as an increased credit score. The input integer variables of the first neural network may include values representing a resource allocation being above a first threshold, within a threshold range, or below a second threshold.”, Paragraph 347, “Some embodiments may determine other types of values, such as a credit score. Some embodiments may determine a plurality of credit scores or other types of scores using a prediction model or set of prediction models based on a value provided via message or a sequence of previous values. Some embodiments may then determine a measure of central tendency based on a plurality of credit scores to determine whether a numeric threshold is satisfied based on the measure of central tendency.”
Hines teaches a neural network model that receives input of a sequence of previous resource-allocation values and a value provided via a message and outputs a credit score for the corresponding record. The model outputs a plurality of credit scores, which corresponds to obtaining a credit score S of the prediction result.).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date to combine the predicted result of the tracking method taught by Fu in view of Zhang in view of Liu’s teaching with Hines’s credit allocation network that generates credit scores in order to evaluate the prediction result against a target objective value reduce training of the neural network model (Paragraph 307 of Hines).
Regarding claim 2,
Fu does not teach wherein in step 1, the … network comprises three convolution layers, two full connection layers and a secondary classification layer; the convolution layers are fixed by offline-training parameters, and the full connection layers and the secondary classification layer are initialized by 500 positive samples and 2000 negative samples.
Zhang, in the same field of endeavor, teaches step 1, the … network comprises three convolution layers, two full connection layers and a secondary classification layer (Page 3 S1.1, “using three-layer convolution layer, two full connection layers and a network structure composed of two classification layers;”);
the convolution layers are fixed by offline-training parameters, and the full connection layers and the secondary classification layer are initialized by 500 positive samples and 2000 negative samples (Page 3 S1.1, “Further, in step S1, positive sample 500, negative sample 2000… the network model through off-line training, fixing the front three layers of convolution layer, only updating the full connection layer and the second classification layer in the tracking process; extracting the characteristic of positive and negative samples in the first frame by the memory network…”
Zhang’s identifying enhanced memory model is built from three convolution layers, two full connection layers, and a two-classification layer, where the convolution layers are fixed by offline training, the full connection layer and second classification layer are initialized from 500 positive and 2000 negative samples.).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date to combine Fu’s space-time memory tracker with Zhang’s network architecture in order to tune the training samples in a way that provides more accuracy when tracking results for the target frame (Page 5 Figure 3 Description of Zhang).
Fu in view of Zhang in further view of Liu does not teach a credit allocation network.
Hines, in the same field of endeavor, teaches a credit allocation network (Paragraph 339, “Using a transformer neural network to predict a schedule may include the timestamps associated with scalar values (e.g., timestamps indicating when a credit utilization changed, timestamps indicating when an amount of a resource was consumed or allocated, etc.) into a corresponding set of time embeddings.”, Paragraph 344, “The first neural network model may be trained to increase the likelihood that the partial assignments will produce a greater objective value, where the objective value may correspond with a target value such as an increased credit score. The input integer variables of the first neural network may include values representing a resource allocation being above a first threshold, within a threshold range, or below a second threshold.”
Hines teaches a neural network model that receives input of a sequence of previous resource-allocation values and a value provided via a message and outputs a credit score for the corresponding record. The model outputs a plurality of credit scores, which corresponds to obtaining a credit score S of the prediction result.).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date to combine the predicted result of the tracking method taught by Fu in view of Zhang in view of Liu’s teaching with Hines’s credit allocation network that generates credit scores in order to evaluate the prediction result against a target objective value reduce training of the neural network model (Paragraph 307 of Hines).
Regarding claim 4,
Fu teaches wherein in step 4, the process of obtaining the location information feature map comprises: calculating a similarity between each pixel of the memory feature f^m and the current frame feature f^c to obtain a similarity matrix Λ, wherein each element of the similarity matrix Λ is calculated according to:
PNG
media_image11.png
85
340
media_image11.png
Greyscale
wherein, i is an index of each pixel on a memory feature map, j is an index of each pixel on the current frame feature map, ⊛ represents a dot product of vectors, and √C is a scale factor set to prevent a corresponding value from being too large wherein C is a dimension of the feature map (Page 4 Section 3.3, "we first compute the similarities between every pixel of fm and every pixel of fq to obtain a similarity matrix w… Taking one element wij for example, we can formally de note wij as:
PNG
media_image12.png
93
391
media_image12.png
Greyscale
where i is the index of each pixel on fm ∈ RTHW×C, j is the index of each pixel on fq ∈ RC×HW, and the binary operator ⊙ denotes vector dot-product. Here s is a scaling factor to prevent the exp function from overflow ing numerically… we set s to √C, where C is the feature dimensionality of fm.”
Fu computes a pixel-wise similarity matrix between the memory feature and the query feature, normalizes it, and scales the dot products by sqrt(C).)
normalizing the similarity matrix Λ by a function softmax, and multiplying the normalized similarity matrix by the memory features to adaptively read the target information stored therein, and stitching the current frame feature and a read-out feature along the channel domain to obtain the location information feature map:
PNG
media_image13.png
39
382
media_image13.png
Greyscale
wherein (·)^T is an matrix transpose operation, and concat(·, ·) represents a series operation along a channel dimension (Page 4 Section 3.3, "we normalize w with a softmax function… Then, treating w as a soft weight map, we multiply fm by w. Because fm stores all historical memory information related to the target, according to the needs of the query frame itself, the target information stored in fm is adaptively retrieved. Obviously, the readout information is a feature map as the same size as fq. Therefore, we concatenate the readout information and the query feature map fq along the channel dimension to generate the final synthetic feature map y. Formally, for the i-th element of y, the space-time memory read operation can be denoted as:
PNG
media_image14.png
38
379
media_image14.png
Greyscale
"
Fu normalizes the similarity matrix with a softmax, multiplies the transposed memory feature by the soft-weight map to read out target information, and concatenates the read-out with the query feature along the channel dimension.)
Regarding claim 6,
Fu does not teach [a]n electronic apparatus having a memory, a processor, and a computer program stored in the memory and capable of running in the processor, wherein, when being executed by the processor, the computer program implements the steps of the single-target tracking method of claim 1.
Zhang, in the same field of endeavor, teaches [a]n electronic apparatus having a memory, a processor, and a computer program stored in the memory and capable of running in the processor, wherein, when being executed by the processor, the computer program implements the steps of the single-target tracking method of claim 1 (Page 7 Paragraph 3 of Zhang, “The example of the invention is realized by using Python on the computer of Intel i7-9700K CPU (3.60GHZ), 16GB RAM and NVDIA Quadro RTX4000. As shown in FIG. 2, identifying the enhanced memory metric performance influence of the memory model. As shown in FIG. 3, an embodiment of the present invention provides a target tracking method based on identifying enhanced memory and intermittent space-time constraints (our) and other classical target tracking algorithm…”
Zhang teaches its target tracking method as a Python program executed on a computer having an Intel i7-9700K CPU and 16 GB of RAM. The recited CPU corresponds to the claimed processor, the 16 GB of RAM corresponds to the claimed memory, and the executable Python program implementing the tracking method corresponds to the computer program that, when executed by the processor, implements the steps of the method. The underlying single-target tracking method of claim 11 is taught by Hu in view of Zhang and in further view of Liu as set forth in claim 1.).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date to combine the tracking method taught by Fu’s teaching with Zhang’s implementation method on a general-purpose computer having a processor and memory in order to carry out the tracking method using conventional available computing hardware (Page 7, Paragraph 3 of Zhang).
Claim 7 recites identical steps to claim 2. Therefore, claim 7 is rejected using the same rationale as claim 2.
Claim 9 recites identical steps to claim 4. Therefore, claim 9 is rejected using the same rationale as claim 4.
Regarding claim 11,
Fu does not teach [a] computer-readable storage medium with a computer program stored thereon, wherein, when being executed by the processor, the computer program implements the steps of the single-target tracking method of claim 1.
Zhang, in the same field of endeavor, teaches [a] computer-readable storage medium with a computer program stored thereon, wherein, when being executed by the processor, the computer program implements the steps of the single-target tracking method of claim 1 (Page 7 Paragraph 3 of Zhang, “The example of the invention is realized by using Python on the computer of Intel i7-9700K CPU (3.60GHZ), 16GB RAM and NVDIA Quadro RTX4000. As shown in FIG. 2, identifying the enhanced memory metric performance influence of the memory model. As shown in FIG. 3, an embodiment of the present invention provides a target tracking method based on identifying enhanced memory and intermittent space-time constraints (our) and other classical target tracking algorithm…”
Zhang teaches its target tracking method as a Python program executed on a computer having an Intel i7-9700K CPU and 16 GB of RAM. The recited CPU corresponds to the claimed processor, the 16 GB of RAM corresponds to the claimed memory, and the executable Python program implementing the tracking method corresponds to the computer program that, when executed by the processor, implements the steps of the method. The underlying single-target tracking method of claim 11 is taught by Hu in view of Zhang and in further view of Liu as set forth in claim 1.).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date to combine the tracking method taught by Fu’s teaching with Zhang’s implementation method on a general-purpose computer having a processor and memory in order to carry out the tracking method using conventional available computing hardware (Page 7, Paragraph 3 of Zhang).
Claim 12 recites identical steps to claim 2. Therefore, claim 12 is rejected using the same rationale as claim 2.
Claim 14 recites identical steps to claim 4. Therefore, claim 14 is rejected using the same rationale as claim 4.
Claims 3, 8, and 13 are rejected under 35 U.S.C. 103 as being unpatentable over Fu (“STMTrack: Template-free Visual Tracking with Space-time Memory Networks”, 2021) in view of Zhang (CN 115272401 A), in view of Liu (“Learning Quality-aware Dynamic Memory for Video Object Segmentation”, 2022), in view of Hines (US 20220075782 A1), and in further view of Lin (“Focal Loss for Dense Object Detection”, 2017).
Regarding claim 3,
Fu in view of Zhang in further view of Liu does not teach the guiding focus loss function is expressed as:
PNG
media_image15.png
98
395
media_image15.png
Greyscale
wherein P is a predicted output, y =1 represents a positive sample, t is the number of iterations, and A is an initial focusing factor.
Lin, in the same field of endeavor, teaches the guiding focus loss function is expressed as:
PNG
media_image15.png
98
395
media_image15.png
Greyscale
wherein P is a predicted output, y =1 represents a positive sample, t is the number of iterations, and A is an initial focusing factor (Page 3 Section 3, “The Focal Loss is designed to address the one-stage object detection scenario in which there is an extreme imbalance between foreground and background classes during training (e.g., 1:1000). We introduce the focal loss starting from the cross entropy (CE) loss for binary classification:
PNG
media_image16.png
64
435
media_image16.png
Greyscale
”, Page 3 Section 3.2, “we propose to add a modulating factor (1 − pt)γ to the cross entropy loss, with tunable focusing parameter γ ≥ 0. We define the focal loss as:
PNG
media_image17.png
47
384
media_image17.png
Greyscale
… The focusing parameter γ smoothly adjusts the rate at which easy examples are down weighted…. In practice we use an α-balanced variant of the focal loss:
PNG
media_image18.png
41
410
media_image18.png
Greyscale
”).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date to combine Fu in view of Zhang in further view of Liu’s teaching with Lin's two-branch focal-loss focal loss structure in order to improve object detection class imbalance during training (Section 3 of Lin).
Claim 8 recites similar limitations to claim 3. Therefore, claim 8 is rejected using the same rationale as claim 3.
Claim 13 recites similar limitations to claim 3. Therefore, claim 13 is rejected using the same rationale as claim 3.
Allowable Subject Matter
Claims 5, 10, and 15 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.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to MAJD MAHER HADDAD whose telephone number is (571)272-2265. The examiner can normally be reached Mon-Friday 8-5 pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Kamran Afshar, can be reached at (571) 272-7796. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/M.M.H./Examiner, Art Unit 2125
/KAMRAN AFSHAR/Supervisory Patent Examiner, Art Unit 2125
Prosecution Insights