DETAILED ACTION
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 .
Information Disclosure Statement
The information disclosure statement(s) (IDS) submitted on February 14th, 2025; August 6th, 2025; February 27th, 2026; June 5th, 2026; June 12th, 2026; and September 17th, 2026, has been considered and the listed references were noted. Due to the extensive number of references provided, the examiner considered only the most pertinent references in favor of efficiency and compact prosecution.
Drawings
The drawings are objected to as failing to comply with 37 CFR 1.84(p)(5) because they do not include the following reference sign(s) mentioned in the description: 321, 322, and 323.
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. 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.
The drawings are objected to as failing to comply with 37 CFR 1.84(p)(5) because they include the following reference character(s) not mentioned in the description: 301 (In Figure 3, 112 is shown instead of 301).
Corrected drawing sheets in compliance with 37 CFR 1.121(d), or amendment to the specification to add the reference character(s) in the description in compliance with 37 CFR 1.121(b) 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. 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
Applicant is reminded of the proper Form of Claims.
The claim or claims must commence on a separate physical sheet or electronic page and should appear after the detailed description of the invention. Any sheet including a claim or portion of a claim may not contain any other parts of the application or other material. While there is no set statutory form for claims, the present Office practice is to insist that each claim must be the object of a sentence starting with "I (or we) claim," "The invention claimed is" (or the equivalent). If, at the time of allowance, the quoted terminology is not present, it is inserted by the Office of Data Management. Each claim begins with a capital letter and ends with a period. Periods may not be used elsewhere in the claims except for abbreviations. See Fressola v. Manbeck, 36 USPQ2d 1211 (D.D.C. 1995). Where a claim sets forth a plurality of elements or steps, each element or step of the claim should be separated by a line indentation, 37 CFR 1.75(i).
There may be plural indentations to further segregate subcombinations or related steps. In general, the printed patent copies will follow the format used but printing difficulties or expense may prevent the duplication of unduly complex claim formats.
See MPEP § 608.01(m) for guidelines for the preparation of formatting the claims.
The claims are objected to because they are center aligned, making it difficult to view the preamble and claim limitations for each claim. The examiner recommends left-aligning all the claims in the claim set and indenting the limitations where appropriate to make it more readable.
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.
The claims in this application are given their broadest reasonable interpretation using the plain meaning of the claim language in light of the specification as it would be understood by one of ordinary skill in the art. The broadest reasonable interpretation of a claim element (also commonly referred to as a claim limitation) is limited by the description in the specification when 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is invoked.
As explained in MPEP § 2181, subsection I, claim limitations that meet the following three-prong test will be interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph:
(A) the claim limitation uses the term “means” or “step” or a term used as a substitute for “means” that is a generic placeholder (also called a nonce term or a non-structural term having no specific structural meaning) for performing the claimed function;
(B) the term “means” or “step” or the generic placeholder is modified by functional language, typically, but not always linked by the transition word “for” (e.g., “means for”) or another linking word or phrase, such as “configured to” or “so that”; and
(C) the term “means” or “step” or the generic placeholder is not modified by sufficient structure, material, or acts for performing the claimed function.
Use of the word “means” (or “step”) in a claim with functional language creates a rebuttable presumption that the claim limitation is to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites sufficient structure, material, or acts to entirely perform the recited function.
Absence of the word “means” (or “step”) in a claim creates a rebuttable presumption that the claim limitation is not to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is not interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites function without reciting sufficient structure, material or acts to entirely perform the recited function.
Claim limitations in this application that use the word “means” (or “step”) are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action. Conversely, claim limitations in this application that do not use the word “means” (or “step”) are not being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action.
This application includes one or more claim limitations that do not use the word “means,” but are nonetheless being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, because the claim limitation(s) uses a generic placeholder that is coupled with functional language without reciting sufficient structure to perform the recited function and the generic placeholder is not preceded by a structural modifier. Such claim limitation(s) is/are: …at least one processing unit… in claim 11.
Because this/these claim limitation(s) is/are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, it/they is/are being interpreted to cover the corresponding structure described in the specification as performing the claimed function, and equivalents thereof.
If applicant does not intend to have this/these limitation(s) interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, applicant may: (1) amend the claim limitation(s) to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph (e.g., by reciting sufficient structure to perform the claimed function); or (2) present a sufficient showing that the claim limitation(s) recite(s) sufficient structure to perform the claimed function so as to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph.
Claim Rejections - 35 USC § 103
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, 3-5, 7, 10, 11, 13-15, 17, and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Anderson et al. (Bottom-Up and Top-Down Attention for Image Captioning and Visual Question Answering) and Chen et al. (CN 115359309 A).
Regarding Claim 1, Anderson discloses (Anderson, pg. 2 (Section 3.1), discloses the following:
PNG
media_image1.png
162
378
media_image1.png
Greyscale
); “respective visual feature representations of at least one object area in the target image” (Anderson, pg. 3 (section 3.1), discloses the following:
PNG
media_image2.png
53
379
media_image2.png
Greyscale
); and “generating, by using a language model, a group of text sequences based on the group of visual feature representations” (Anderson, pg. 3 (Section 3.1), discloses the following:
PNG
media_image3.png
84
371
media_image3.png
Greyscale
As one of ordinary skill in the art would know, captioning is considered a group of text sequences, as it describes what is being seen in the image); “each text sequence indicating at least one category to which an object in an object area corresponding to the visual feature representation belongs” (Anderson, Figure 5, discloses the following:
PNG
media_image4.png
567
769
media_image4.png
Greyscale
As described by the caption in Figure 5, the text sequences are categorized by the specific object and what that object is doing when the image is acquired. We can also see how the caption corresponds to the specific area with the image relating to that caption, making it more accurate when performing image processing using the algorithm Anderson is proposing). Therefore, it would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to use the techniques of extracting visual features and generating text sequences for the visual features within the image seen in Anderson to improve the object detection method in the same way. By using the techniques seen in Anderson, one of ordinary skill in the art allows for a user to effectively detect objects within the image while simultaneously being able to describe the object's state. Thus, it would have been obvious for one of ordinary skill in the art to use the Anderson reference to achieve the above-described limitations described in Claim 1.
Anderson does not explicitly disclose “An object detection method, comprising”. However, in an analogous field of endeavor, Chen discloses “The invention provides a target detection model training method and device, equipment and a medium, relates to the field of artificial intelligence, in particular to the technical field of deep learning, image processing and computer vision, and can be applied to scenes such as face recognition” (Chen, Abstract). It is important to know that target detection model training method is synonymous with an object detection method. Therefore, it would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to combine the object detection model seen in Anderson with the object detection method seen in Chen to achieve a complete object detection method, as seen in Claim 1.
Regarding Claim 3, the combination of Anderson and Chen discloses “The method according to claim 1, wherein the language model outputs a group of output sequences by: generating, for each visual feature representation, a text sequence corresponding to the visual feature representation based on a semantic correlation between the visual feature representation and a text feature representation of a category” (Anderson, pg. 4 (Section 3.2.2), discloses the following:
PNG
media_image5.png
641
375
media_image5.png
Greyscale
As shown, visual features from the image are used in the LSTM to determine a group of text sequences corresponding to each visual feature, as shown in Figure 3. Although this excerpt does not directly mention the semantic correlation between the visual feature representation and text feature, it can be interpreted by one of ordinary skill in the art that the text would be semantically correlated to the visual feature due to the fact that the Top-Down Attention LSTM tracks the generated caption context up to the current time step. Additionally, the attend block computes the correlation between the visual features and the text via the learned weights, thus aligning the specific words with their visual objects to then allow the Language LSTM to use this context to determine the probability over the next possible set of output words.).
Regarding Claim 4, the combination of Anderson and Chen discloses “The method according to claim 1, wherein generating, by using the language model, the group of text sequences based on the group of visual feature representations comprises: performing, by using a visual-language feature adapter associated with the language model, a conversion on the group of visual feature representations to obtain a group of converted feature representations” (Anderson, pg. 4, discloses the following:
PNG
media_image6.png
622
380
media_image6.png
Greyscale
As shown, the visual features (v) are being used to develop a new group of features via the convex combination of all input features (5)); and “obtaining, by providing the group of converted feature representations to the language model, the group of text sequences generated by the language model” (Anderson, pg. 4 (Section 3.2.2), discloses the following:
PNG
media_image5.png
641
375
media_image5.png
Greyscale
As shown here, the input from the top-down attention LSTM is being input into the Language LSTM model to generate a group of output text sequences, as seen in Figure 3).
Regarding Claim 5, the combination of Anderson and Chen discloses “The method according to claim 4, wherein the language model and the visual-language feature adapter are trained, and during a training process of the object detection model, parameters of the trained language model and the trained visual-language feature adapter are fixed” (Anderson, pg. 5 and 6, disclose the following:
PNG
media_image7.png
502
381
media_image7.png
Greyscale
It is important to know that the visual-language feature adapter can be interpreted as the Faster R-CNN. This is because, as mentioned in the analysis of Claim 1, the Faster R-CNN is considered an object detection model, which in turn extracts features from an image. Additionally, we can evidently see both the Faster R-CNN and language model (LSTM) being trained through a Visual Genome dataset, whereas the parameters as mentioned in the last paragraph are fixed throughout the experimentation process).
Regarding Claim 7, the combination of Anderson and Chen discloses “The method according to claim 1, wherein the object detection model and the language model are jointly trained” (Anderson, pg. 5-6 (Section 3.4), disclose the following:
PNG
media_image8.png
784
595
media_image8.png
Greyscale
As explained in the analysis of Claim 1 and 5, Faster R-CNN is an object detection model, and the language model is the LSTM. We can see evidently in Section 3.4 that both models are being trained with the Visual Genome model, which in turn shows that they are jointly trained).
Regarding Claim 10, the combination of Anderson and Chen discloses “The method according to claim 1, further comprising: providing an object detection result for the target image, the object detection result indicating the at least one object area that is located from the target object by the object detection model and a text sequence corresponding to each object area that is generated by the language model” (Anderson, pg. 2 (Section 3.1), discloses the following:
PNG
media_image9.png
1078
372
media_image9.png
Greyscale
As shown in Section 3.1, the object detection result is described in Figure 2, where we can see the various objects being detected within the image with labels associated with them; Anderson, pg. 4 (Section 3.4), discloses the following:
PNG
media_image5.png
641
375
media_image5.png
Greyscale
Here in Section 3.2.2, the language LSTM serves as the fundamental building block for the text sequences seen within Figure 2, making this correlate with the final object detection result).
Claim 11 recites an apparatus (electronic device) with features corresponding to the elements of the method recited in Claim 1. Therefore, the recited features of this claim are mapped to the proposed combination in the same manner as the corresponding elements in its corresponding method claim. Additionally, the rationale and motivation to combine Anderson and Chen references, presented in rejection of Claim 1, apply to this claim. Finally, the combination of Anderson and Chen discloses an electronic device with a processing unit coupled to memory (For example, see Chen, Paragraph [0010], where they disclose “According to another aspect of the present disclosure, there is provided an electronic device including: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform a method of training or a method of object detection of the above-described object detection model”).
Claim 13 recites an apparatus with features corresponding to the elements of the method recited in Claim 3. Therefore, the recited features of this claim are mapped to the proposed combination in the same manner as the corresponding elements in its corresponding method claim. Additionally, the rationale and motivation to combine the Anderson and Chen references, presented in rejection of Claim 3, apply to this claim.
Claim 14 recites an apparatus with features corresponding to the elements of the method recited in Claim 4. Therefore, the recited features of this claim are mapped to the proposed combination in the same manner as the corresponding elements in its corresponding method claim. Additionally, the rationale and motivation to combine the Anderson and Chen references, presented in rejection of Claim 4, apply to this claim.
Claim 15 recites an apparatus with features corresponding to the elements of the method recited in Claim 5. Therefore, the recited features of this claim are mapped to the proposed combination in the same manner as the corresponding elements in its corresponding method claim. Additionally, the rationale and motivation to combine the Anderson and Chen references, presented in rejection of Claim 5, apply to this claim.
Claim 17 recites an apparatus with features corresponding to the elements of the method recited in Claim 7. Therefore, the recited features of this claim are mapped to the proposed combination in the same manner as the corresponding elements in its corresponding method claim. Additionally, the rationale and motivation to combine the Anderson and Chen references, presented in rejection of Claim 7, apply to this claim.
Claim 20 recites a computer-readable storage medium storing a program with instructions corresponding to the steps recited in Claim 1. Therefore, the recited programming instructions of this claim are mapped to the proposed combination in the same manner as the corresponding steps in its corresponding method claim. Additionally, the rationale and motivation to combine the Anderson and Chen references, presented in rejection of Claim 1, apply to this claim. Finally, the combination of Anderson and Chen references discloses a computer readable storage medium (for example, see Chen, Paragraph [0011], where they disclose: “According to another aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium storing computer instructions for causing the computer to execute the training method or the object detection method of the above object detection model.”).
Claims 2 and 12 are rejected under 35 U.S.C. 103 as being unpatentable over Anderson in view of Chen, and further in view of Luo et al. (CN 111859005 A) and Chen 2 et al. (US 2022/0067379).
Regarding Claim 2, the combination of Anderson and Chen discloses “The method according to claim 1” (Please refer to the above-described analysis regarding Claim 1), Anderson, pg. 4, Section 3.4, discloses the following:
PNG
media_image10.png
90
364
media_image10.png
Greyscale
) (Luo, Paragraph [0031]). Here, we can see that the output sequence here has a length according to the length of the input sequence, making it predetermined after being found via the object detection model seen in Luo. Therefore, it would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to combine the method seen in the combination of Anderson and Chen with the Luo technique of using an output sequence having a predetermined length to achieve the above-described limitations seen in Claim 2.
The combination of Anderson, Chen, and Luo does not explicitly disclose “the at least one detection result corresponding to each object comprises a predetermined number of detection results that match the predetermined length”. However, in an analogous field of endeavor, Chen 2 discloses “In a possible implementation, to improve the accuracy of the category labelling result, after the detection result of the target video frame in the video stream is determined, a total number of detection results obtained in a preset time interval may also be determined” (Chen 2, Paragraph [0049]). It is important to note that due to the number of detection results being determined by a specific time interval, this in turn shows that the detection results are within the predetermined length, whereas the length is the preset time interval that each result is within. Therefore, it would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to combine the method seen in the combination of Anderson, Chen, and Luo with the Chen 2 technique of having a predetermined number of detection results that match a predetermined length to achieve an improved object detection method, as seen in the method of Claim 2.
Claim 12 recites an apparatus with features corresponding to the elements of the method recited in Claim 2. Therefore, the recited features of this claim are mapped to the proposed combination in the same manner as the corresponding elements in its corresponding method claim. Additionally, the rationale and motivation to combine the Anderson, Chen, Luo, and Chen 2 references, presented in rejection of Claim 2, apply to this claim.
Claims 6 and 16 are rejected under 35 U.S.C. 103 as being unpatentable over Anderson in view of Chen, and further in view of Liu et al. (Chinese Image Caption Generation via Visual Attention and Topic Modeling).
Regarding Claim 6, the combination of Anderson and Chen discloses “The method according to claim 5, wherein the object detection model comprises” (Please refer to the above-described analysis regarding Claim 5) pg. 1250-1251:
PNG
media_image11.png
1065
389
media_image11.png
Greyscale
We can see from the following section that the network contains an image encoder-decoder model, which is trained via the ImageNet database. The parameters are interpreted as the image feature dimension and the region size, which evidently stay fixed when the network is being trained with multiple images. Therefore, it would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to combine the object detection method seen in the combination of Anderson and Chen with the Liu technique of having an image encoder-decoder model that is trained with fixed parameters to achieve a more improved object detection method. By combining the object detection method seen in the combination of Anderson and Chen with the image encoder-decoder model seen in Liu, one of ordinary skill in the art allows the user to easily extract features and decipher what is important within the image to accurately detect objects using language models. Thus, it would have been obvious for one of ordinary skill in the art to combine the Anderson, Chen, and Liu references to achieve the same method seen in Claim 6.
Claim 16 recites an apparatus with features corresponding to the elements of the method recited in Claim 6. Therefore, the recited features of this claim are mapped to the proposed combination in the same manner as the corresponding elements in its corresponding method claim. Additionally, the rationale and motivation to combine the Anderson, Chen, and Liu references, presented in rejection of Claim 6, apply to this claim.
Claims 8, 9, 18, and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Anderson in view of Chen, and further in view of Han et al. (CN 117251592 A).
Regarding Claim 8, the combination of Anderson and Chen discloses “The method according to claim 1” (Han, Paragraph [0185]). Han further discloses the following in Paragraphs [0188]- [0190]:
PNG
media_image12.png
297
652
media_image12.png
Greyscale
In the above paragraphs, we can clearly see that the processing result generated by the training sample of the image encoder can be interpreted as obtaining the first training dataset. The processing result is derived from the image sample with the category and plurality of detection frames found within the result, as seen in Paragraph [0189]. Additionally, the ground truth position in this case can be defined as true value, which is used alongside the processing result to update the image encoder and decoder within the object detection model. We can also consider the candidate detection boxes as a way of determining the estimated position alongside the category within the first processing result being the first estimated category. It is important to note that the distance loss and semantic loss are implicitly described here in Paragraphs [0189] and [0190], as the semantic loss can be calculated by comparing the image encoder's output (the category of each detection frame) against the corresponding true value while the distance loss can be calculated by comparing the predicted localization coordinates (plurality of detection frames) against their corresponding true value. Therefore, it would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to combine the method seen in the combination of Anderson and Chen with the Han techniques of obtaining a training dataset, determining the position and category for each object, and updating the object detection model to improve the object detection method in the same way.
Regarding Claim 9, the combination of Anderson Chen, and Han discloses “The method according to claim 8, wherein at least the object detection model is trained by: obtaining, by providing the first sample image to the updated object detection model and the language model, a pseudo-label category output by the language model, the pseudo-label category being different from the sample category” (Chen, Paragraph [0006], discloses “a training method of a target detection model, including: acquiring a first sample image, and labeling the real category of a target object included in the first sample image; inputting the first sample image into a target detection model to obtain a first feature vector representation of the target object output by the target detection model and a second feature vector representation of an object in the first sample image corresponding to each of a plurality of initial prediction categories”; Paragraph [0057], discloses “As shown in fig. 3, according to some embodiments, the target detection model includes a visual language model and a position regressor, the visual language model includes an image feature extraction sub-model and a transformation matrix...”; From Paragraphs [0006] and [0057], we see that the pseudo-label category is synonymous to the initial prediction category, and that the second feature vector representation of an object corresponding to each of the initial prediction categories is generated by the target detection model, which includes a visual language model. Due to the feature vectors representing different objects, this means that the initial prediction categories for each vector are different); “determining, by providing the first sample image to the updated object detection model and the language model, second estimated position information and a second estimated category of the object area in the first sample image” (Chen, Paragraph [0057], discloses “… step S202 inputs the first sample image into the target detection model to obtain a first feature vector representation and predicted position information of the target object output by the target detection model, and the second feature vector representation of the object corresponding to each of a plurality of initial prediction categories in the first sample image includes: inputting the first sample image into an image feature extraction sub-model to obtain an image feature vector representation of the target object output by the image feature extraction sub-model and an image feature vector representation of an object corresponding to each of a plurality of initial prediction categories in the first sample image; inputting the image feature vector representation of the target object into the position regressor to obtain the predicted position information output by the position regressor; and inputting the image characteristic vector representation of the target object and the image characteristic vector representation of the object corresponding to the plurality of initial prediction classes into a conversion matrix to obtain a first characteristic vector representation of the target object and a second characteristic vector representation of the object corresponding to each initial prediction class output by the conversion matrix. Therefore, the image feature vector representation of each object in the sample image can be obtained by using the visual language model and then projected into the corresponding text feature vector …”; Paragraph [0058], discloses “Illustratively, the visual language model may be a pre-trained visual language model, which may for example comprise the steps of: acquiring a pre-training sample image and real class text labels of objects included in the pre-training sample image; inputting the real category text labels into a pre-trained text model, the pre-trained text model being configured to map text information into a text feature vector representation to obtain a corresponding real category text feature vector representation output by the pre-trained text model; inputting the pre-training sample image into an initial visual language model to obtain a prediction class image feature vector representation of an object included in the sample image output by the initial visual language model; inputting the prediction category image feature vector representation into a conversion matrix to obtain a prediction category text feature vector representation output by the conversion matrix, wherein the prediction category text feature vector representation and the real category text feature vector representation are in the same vector space”; As we can see in the aforementioned paragraphs, the second estimated position is determined via inputting the image feature vector into the position regressor, while the second estimated category (prediction category) is found via inputting the sample image into the visual language model, which is a part of the target detection model.); and “updating the object detection model based on a second distance loss between the second estimated position information and the ground-truth position information” (Chen, Paragraph [0056], discloses “Further, the method may further include a training method for position information of the target object. According to some embodiments, the method further comprises: marking real position information of a target object included in the first sample image; acquiring predicted position information of the target object output by the target detection model based on a first sample image; calculating a third loss value based on the real position information and the predicted position information of the target object, and wherein the target detection model is parametrized based on the first loss value, the second loss value, and the third loss value. Therefore, by applying the training method for the position information of the target object in the training process of the target detection model, the target detection model can simultaneously output the position information of the target object and the feature information indicating the type of the target object based on the input image, and end-to-end target detection is realized”; Here, the real position is synonymous to the ground truth position, as it signifies the object's actual position within the image), and “a second semantic loss between the sample category” (Chen, Paragraph [0058], discloses “calculating a loss value based on the prediction category text feature vector representation and the true category text feature vector representation; adjusting parameters of the initial visual language model based on the loss value. By executing the pre-training operation, a pre-training visual language model with certain visual recognition capability can be obtained”; Similar to Paragraph [0056], the true category is synonymous to the sample category); and “the second estimated category and a second semantic loss between the pseudo-label category and the second estimated category” (Chen, Paragraph [0008], discloses “a second determining unit configured to determine a true prediction class from the plurality of initial prediction classes based on respective confidence degrees of the plurality of initial prediction classes, and label an object of the first sample image including the true prediction class; a second calculation unit configured to calculate a second loss value based on the confidence degree corresponding to the true prediction category”; Here, the pseudo-label category is interpreted as the initial prediction class, and the second estimated category is interpreted as the true category).
Claim 18 recites an apparatus with features corresponding to the elements of the method recited in Claim 8. Therefore, the recited features of this claim are mapped to the proposed combination in the same manner as the corresponding elements in its corresponding method claim. Additionally, the rationale and motivation to combine the Anderson, Chen, and Han references, presented in rejection of Claim 8, apply to this claim.
Claim 19 recites an apparatus with features corresponding to the elements of the method recited in Claim 9. Therefore, the recited features of this claim are mapped to the proposed combination in the same manner as the corresponding elements in its corresponding method claim. Additionally, the rationale and motivation to combine the Anderson, Chen, and Han references, presented in rejection of Claim 9, apply to this claim.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Mao et al. (US 2025/0104453 A1, w/ an EFD of March 21st, 2022) teaches an image description generation method and apparatus, a device, a medium, and a product, and relates to the technical field of image processing
Mao 2 et al. (US 2025/0095327, w/ an EFD of January 21st, 2022) teaches a method, apparatus, readable storage medium, and electronic device for object attribute recognition.
Miao et al. (CN 113837257 A) teaches a training method of target detection model and target detection method, target detection, not only based on the image itself, but also the description text of the image as the consideration factor of the target detection.
Yue et al. (US 20230020965 A1) teaches a method and apparatus for updating an object recognition model in the field of artificial intelligence.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to SORIE I KOROMA JR whose telephone number is (571)272-9259. The examiner can normally be reached Monday - Friday 8AM-6:00PM; Alternate Fridays Off.
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, Amandeep Saini can be reached at 571-272-3382. 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.
/SORIE I KOROMA JR/Examiner, Art Unit 2662
/Siamak Harandi/Primary Examiner, Art Unit 2662
Prosecution Insights