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 .
Response to Arguments
Applicant’s arguments and amendment have persuasively overcome some of the 112 rejections.
The remaining issues are addressed below.
Title
Applicant argues:
Applicant respectfully traverses this objection. The title of the application is "Training Models for Object Detection," which is clearly indicative of the claimed subject matter.
Examiner responds:
"Training Models for Object Detection" is too generic to be a useful title.
Applicant argues:
The recited "processor resource" would be readily understood by a person of ordinary skill in the art, particularly in light of the Specification (see, e.g., [0039]).
Examiner responds:
Specification [0039] explains “processor resource” differently than [0046] does. Additionally, both paragraphs repeatedly say “may be” and thus it is incorrect to limit the claim to examples offered as “may be.”
Applicant argues:
The recited "non-transitory memory resource" would be readily understood by a person of ordinary skill in the art, particularly in light of the Specification (see, e.g., [0040]).
Examiner responds:
Specification [0040] states that the “memory resource" can be remote, but the claim states that the memory resource is comprised in a computing device (i.e., the memory resource is local, not remote). Additionally, Specification [0040] repeatedly says “may be” and thus it is incorrect to limit the claim to examples offered as “may be.”
Applicant argues:
The Office's interpretation is not reasonable, by their plain meaning and also in light of the specification.
Examiner responds:
The claim does not specify that the processor resource is performing these steps, thus the examiner believes that the processor resource asking a user to perform these steps on a different computer meets the claim limitations.
Applicant argues:
The recited "object" would be readily understood by a person of ordinary skill in the art in light of the Specification.
Examiner responds:
This does not resolve the question posed in the rejection. Here, a citation to the specification would help the examiner understand Applicant’s perspective.
Applicant argues:
The recited "a category of objects" would be readily understood by a person of ordinary skill in the art in light of the Specification.
Examiner responds:
The examiner does not understand how Applicant believes one of ordinary skill in the art would understand this. Applicant may wish to submit evidence, such as a declaration.
101
Applicant argues:
Claim 1 does not recite a mental process under Step 2A, Prong One.
Examiner responds:
The recited steps are generic. As both the guidance cited in the office action and Recentive Analytics, Inc. v. Fox Corp., 134 F. 4th 1205 (Fed. Cir. 2025) direct, generic use of artificial intelligence is a mental process.
Applicant argues:
Here, claim 1 recites features reflecting improvements in training a machine learning model.
Examiner responds:
Supervised learning was well-understood, routine and conventional as of Applicant’s effective date. See, e.g., https://en.wikipedia.org/w/index.php?title=Convolutional_neural_network&oldid=1033189743 stating “Several supervised and unsupervised learning algorithms have been proposed over the decades to train the weights of a neocognitron.[9] Today, however, the CNN architecture is usually trained through backpropagation.” Reference 9 is: Fukushima, K. (2007). "Neocognitron". Scholarpedia. 2 (1): 1717. Bibcode:2007SchpJ...2.1717F. doi:10.4249/scholarpedia.1717.
Applicant argues:
Much like in Desjardins and Carmody, claim 1 of the present application recites features that reflect improvements in training a machine learning model, and thus is patent-eligible.
Examiner responds:
Both Desjardins and Carmody allow the fact finder to consider the state of the art, as is apparent from the phrase “improvements in training a machine learning model” (i.e., one needs to know what is currently done to understand whether or not something is an improvement).
Applicant argues:
Utilizing inferencing and further training, the CNN model can be revised to improve its object detection accuracy. Accordingly, such an approach can provide an accurate object detector with a lower error rate than previous approaches …
Examiner responds:
Applicant may wish to submit evidence comparing the current technique with the then state of the art. If Applicant shows an improvement in training over the state of the art, this is expected to overcome the 101 rejection.
As to the prior art, a new mapping has been provided that more directly speaks to the amended claims
Specification
The title of the invention is not descriptive. A new title is required that is clearly indicative of the invention to which the claims are directed.
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 1-17 (all claims) 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.
Claim 1 recites “a processor resource,” but this is new terminology. MPEP 2173.05(a). Reciting “a processor” is expected to overcome this rejection. If Applicant wishes to identify an electronic circuit (as per specification [0039]) that is a processor resource but not a processor, the examiner will consider it.
Claim 1 recites “a non-transitory memory resource,” as above, but this is new terminology. MPEP 2173.05(a).
Claim 1 recites instructions that cause a processor resource to “cause” various actions. The broadest reasonable interpretation of this second cause includes putting a message on a screen asking a user to do the following.
Claims 8 and 12 recite corresponding language that raises the same issue.
Claims 1, 8, and 12 recite “object,” but it is unclear if this is intended to refer to one instance of an object (the plain meaning), or a type of object (as used in, for example, claims 3 and 5).
Claims 9 and 11 recite “a category of objects,” but this is subjective because different people can have different ideas as to what the category is and what constitutes a category. MPEP 2173.05(b)(IV).
Claim 17 recites “corresponding to,” but this is subjective. MPEP 2173.05(b)(IV). The concern is that different people can have different opinions about whether one image corresponds to another (perhaps both are from the same street, perhaps both are of cars). Reciting that the annotated image “is” the unannotated image, but with annotations is expected to overcome this rejection because “is” is an objective standard.
Dependent claims are likewise rejected.
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 1-17 (all claims) are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea (mental process) without significantly more.
Step 1:
Claim 1 (and its dependents) recite a device, and machines satisfy Step 1 of the eligibility test.
Claim 8 (and its dependents) recite a non-transitory computer readable storage medium, and manufactures satisfy Step 1 of the eligibility test.
Claim 12 (and its dependents) recite a method, and processes satisfy Step 1 of the eligibility test.
Step 2A, prong one: All of the elements of claims 1-20 are a mental process because a person can check their work and study harder. Further, the various models are also mental processes, see example 47, claim 2, element (d) (from the July 2024 AI subject matter eligibility examples). MPEP 2106.04(a)(2)(III)(C) explains that use of a generic computer or in a computer environment is still a mental process. In particular, this section begins by citing Gottschalk v. Benson, 409 US 63 (1972). “The Supreme Court recognized this in Benson, determining that a mathematical algorithm for converting binary coded decimal to pure binary within a computer’s shift register was an abstract idea.” In Benson the Supreme Court did not separately analyze the computer hardware at issue; the specifics of what hardware was claimed is only included in an appendix to the decision.
Because there are no additional elements, no further analysis is required for Step 2A, prong two or Step 2B.
Amending the claims to be more technical, such that the analogy to what a human would do is less apparent, is expected to assist in overcoming this rejection.
Claim Rejections - 35 USC § 102
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 the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claims 1-17 (all claims) are rejected under 35 U.S.C. 102(a)(1) and/or (a)(2) as being anticipated by US20200134375A1 (“Zhan”).
1. A computing device, comprising: (Zhan, claim 12, “A semantic segmentation model training apparatus”)
a processor resource; and (Zhan, claim 12, “a processor”)
a non-transitory memory resource storing machine-readable instructions that, when executed, cause the processor resource to: (Zhan, claim 12, “a memory storing processor-executable instructions”)
cause a convolutional neural network (CNN) model to be trained with an initial training data set to detect an object included in annotated images included in the initial training data set; (Zhan, claim 12, “obtaining, by a convolutional neural network based on the category of the at least one unlabeled image and a category of at least one labeled image, sub-images respectively corresponding to at least two images and features corresponding to the sub-images.” See also, [0047] “training the semantic segmentation model by a gradient back propagation algorithm, so as to minimize an error of the convolutional neural network.” Fig. 3 shows that Zhan’s CNN is part of Zhan’s semantic segmentation model.)
cause the trained CNN model to perform inferencing on a plurality of unannotated images included in an inference data set to detect the object in the plurality of unannotated images; (Zhan, claim 12, “obtaining, by a convolutional neural network based on the category of the at least one unlabeled image and a category of at least one labeled image, sub-images respectively corresponding to at least two images and features corresponding to the sub-images”)
determine an error rate of the trained CNN model, wherein the error rate is a ratio between a quantity of unannotated images for which the trained CNN model misdetected the object and a quantity of the plurality of unannotated images; and (Zhan, [0034] “By training, a semantic segmentation model learnt in a self-supervised mode and having relatively strong semantic distinction is obtained, and high accuracy in semantic segmentation can be achieved.” Zhan’s self-supervision teaches the claimed unannotated images, and “accuracy” is understood in its technical sense as teaching the error rate, see, e.g., https://en.wikipedia.org/w/index.php?title=Accuracy_and_precision&oldid=999701614, specifically the section titled “In binary classification”)
cause the trained CNN model to be further trained based on the error rate. (Zhan, claim 16 “iteratively implementing following operations until the maximum error is lower than or equal to a preset value: … .”)
2 The computing device of claim 1, wherein the processor resource is to cause the trained CNN model to be further trained with a revised training data set to revise the CNN model. (Zhan, claim 12, “training the semantic segmentation model on the basis of the categories of the at least two sub-images and feature distances between the at least two sub-images.”)
3. The computing device of claim 2, wherein the revised training data set includes annotated images having objects that were mis-detected during the inferencing on the plurality of unannotated images. (Zhan, claim 15, “training the semantic segmentation model by a gradient back propagation algorithm, so as to minimize an error of the convolutional neural network, wherein the error is a triplet loss of the features of the corresponding sub-images obtained based on the convolutional neural network.” Zhan’s triplet loss teaches the claimed mis-detected.)
4. The computing device of claim 1, wherein the processor resource is to cause the trained CNN model to be further trained in response to the error rate being greater than a threshold amount. (Zhan, claim 16 “iteratively implementing following operations until the maximum error is lower than or equal to a preset value: … .”)
5. The computing device of claim 1,wherein the annotated images in the initial training data set are annotated with bounding boxes around the object. (Zhan, 0062, “outputting the image in the select box as a sub-image, and labeling the sub-image as said category.” Zhan’s output images are considered part of the initial training data because Zhan describes them as part of the initial learning. See, e.g., Zhan, [0075]. Zhan Fig. 2, box 211 shows the select box teaches the claimed bounding box around the object.)
6. The computing device of claim 1, wherein the unannotated images in the inference data set include the object without bounding boxes around the object. (Zhan, Fig. 3.)
7. The computing device of claim 1, wherein the object is a face of a person. (Zhan, Fig. 2, box 211.)
Claim 8 is rejected as per claim 3.
9. The non-transitory machine-readable storage medium of claim 8, wherein the object is included in a category of objects intended for detection. (Zhan, Fig. 2, box 211.)
10. The non-transitory machine-readable storage medium of claim 8, wherein misdetection of the object includes an image included in the inference data set having an object to be detected that was not detected. (Zhan, [0034] “both the labeled image and the unlabeled image are applied to training, thereby achieving self-supervised training.” Zhan’s self-supervised learning teaches the claimed misdetection.)
11. The non-transitory machine-readable storage medium of claim 8, wherein misdetection of the object includes an image included in the inference data set having an object that was detected, but not being of a category of objects intended for detection. (Zhan, [0034] “both the labeled image and the unlabeled image are applied to training, thereby achieving self-supervised training.” Zhan’s self-supervised learning teaches the claimed misdetection.)
Claims 12-14 are rejected as per claim 3.
Claim 15 is rejected as per claim 4.
16. (New) The computing device of claim 1, wherein the machine-readable instructions, when executed, cause the processor resource to determine the error rate of the trained CNN model by:
determining, from the inferencing on the plurality of unannotated images, false-positive detections of the object associated with the plurality of unannotated images or false-negative detections of the object associated with the plurality of unannotated images. (Zhan, [0034] “By training, a semantic segmentation model learnt in a self-supervised mode and having relatively strong semantic distinction is obtained, and high accuracy in semantic segmentation can be achieved.” Zhan’s self-supervision teaches the claimed unannotated images, and “accuracy” is understood in its technical sense as teaching the error rate, see, e.g., https://en.wikipedia.org/w/index.php?title=Accuracy_and_precision&oldid=999701614, specifically the section titled “In binary classification”)
17. (New) The computing device of claim 2, wherein the revised training data set comprises annotated images corresponding to the unannotated images for which the trained CNN model misdetected the object. (Zhan, claim 12, “obtaining, by a convolutional neural network based on the category of the at least one unlabeled image and a category of at least one labeled image, sub-images respectively corresponding to at least two images and features corresponding to the sub-images”)
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to DAVID ORANGE whose telephone number is (571)270-1799. The examiner can normally be reached Mon-Fri, 9-5.
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, Gregory Morse can be reached at 571-272-3838. 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.
/DAVID ORANGE/ Primary Examiner, Art Unit 2663
Prosecution Insights