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 .
Response to Arguments
Applicant’s arguments with respect to claims 81-100 have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
Specification
Applicant is reminded of the proper language and format for an abstract of the disclosure.
The abstract should be in narrative form and generally limited to a single paragraph on a separate sheet within the range of 50 to 150 words. It is important that the abstract not exceed 150 words in length since the space provided for the abstract on the computer tape used by the printer is limited. The form and legal phraseology often used in patent claims, such as "means" and "said," should be avoided. The abstract should describe the disclosure sufficiently to assist readers in deciding whether there is a need for consulting the full patent text for details.
The language should be clear and concise and should not repeat information given in the title. It should avoid using phrases which can be implied, such as, "The disclosure concerns," "The disclosure defined by this invention," "The disclosure describes," etc.
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 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 set forth in Graham v. John Deere Co., 383 U.S. 1, 148 USPQ 459 (1966), that are applied 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.
Claims 81 and 91 are rejected under 35 U.S.C. 103 as being unpatentable over Yuchai et al. (CN112163599A image classification method based on multi-scale and multi-level fusion), hereinafter, “Yuchai”, in view of Mao et al. (USPAP 2020/0202,145), hereinafter, “Mao”, and further in view of Wang et al. (EP 3278559 B1, TRAINING END-TO-END VIDEO PROCESSES), hereinafter, “Wang”.
Regarding claim 81 Yuchai teaches, train or finetune one or more additional parameters of at least one neural network (NN) or a portion of the at least one NN, wherein the one or more additional parameters comprise one or more scaling parameters. (Please note, paragraph 0038. As indicated for training feature fusion classifiers, a transfer learning strategy can be used for parameter learning. That is, first, the neural network parameters are pre-trained on a large-scale existing dataset, and then the neural network is transferred to a specific application domain, where the neural network parameters are fine-tuned using a small amount of data from that domain.).
Yuchai does not expressly teach, encode, decode, or process one or more media elements based on the at least one neural network or the portion of the at least one NN comprising the trained or finetuned one or more additional parameters.
Mao teaches, encode, decode, or process one or more media elements based on the at least one neural network or the portion of the at least one NN comprising the trained or finetuned one or more additional parameters. (Please note, paragraph 0075. As indicated the object classifier neural network processes the patch components of the training example and the selected feature vector in accordance with current values of the parameters of the network. In this regard, examiner considers claimed, “media element”, to correspond to the cited, “patch”.).
Yuchai & Mao are combinable because they are from the same field of endeavor.
At the time before the effective filing date, it would have been obvious to a person of ordinary skill in the art to utilize this processing the patch components of Mao in Yuchai’s invention.
The suggestion/motivation for doing so would have been as indicated on paragraph 0075, “to generate a predicted object classification for the object of interest represented in the object patches.”.
Yuchai and Mao do not expressly recite, during an inference stage when a codec comprising the at least one NN compresses or decompresses one or more media elements and the apparatus comprises the codec or the codec comprises the apparatus.
Wang recites during an inference stage when a codec comprising the at least one NN compresses or decompresses one or more media elements and the apparatus comprises the codec or the codec comprises the apparatus. (Please note, page 5, 5th paragraph. As indicated each scene or segment of an original media file is compressed using one of a plurality of codecs. A particular codec best suited to compressing each scene is selected from a codec library. An Al system can be used to determine whether a codec exists in the library that has previously been found to optimally compress a scene. The Al system can be implemented as a neural network, which can be used to select codec settings, such as preprocessing/postprocessing filters.).
Yuchai, Mao & Wang are combinable because they are from the same field of endeavor.
At the time of the invention, it would have been obvious to a person of ordinary skill in the art to utilize this during an inference stage when a codec comprising the at least one NN compresses or decompresses one or more media elements of Wang in Yuchai & Mao’s invention.
The suggestion/motivation for doing so would have been as indicated on page 5, 5th paragraph, “for optimizing the compression of video and audio signals.”.
Therefore, it would have been obvious to combine Yuchai, Mao with Wang to obtain the invention as specified in claim 81.
Regarding claim 91, analysis similar to those presented for claim 81, are applicable.
Claims 82-83, 89-90, 92-93 and 99 are rejected under 35 U.S.C. 103(a) as being unpatentable over Yuchai et al. (CN112163599A image classification method based on multi-scale and multi-level fusion), hereinafter, “Yuchai”, in view of Mao et al. (USPAP 2020/0202,145), hereinafter, “Mao”, and further in view of Wang et al. (EP 3278559 B1, TRAINING END-TO-END VIDEO PROCESSES), hereinafter, “Wang”, as applied to claims 81 and 91 above, and further in view of Ge et al. (USPAP 2020/0314,827), hereinafter, “Ge”.
Regarding claim 82, Yuchai teaches, train or finetune one or more additional parameters of at least one neural network (NN) or a portion of the at least one NN, wherein the one or more additional parameters comprise one or more scaling parameters. (Please note, paragraph 0038. As indicated for training feature fusion classifiers, a transfer learning strategy can be used for parameter learning. That is, first, the neural network parameters are pre-trained on a large-scale existing dataset, and then the neural network is transferred to a specific application domain, where the neural network parameters are fine-tuned using a small amount of data from that domain.).
Yuchai does not expressly teach, encode, decode, or process one or more media elements based on the at least one neural network or the portion of the at least one NN comprising the trained or finetuned one or more additional parameters.
Mao teaches, encode, decode, or process one or more media elements based on the at least one neural network or the portion of the at least one NN comprising the trained or finetuned one or more additional parameters. (Please note, paragraph 0075. As indicated the object classifier neural network processes the patch components of the training example and the selected feature vector in accordance with current values of the parameters of the network. In this regard, examiner considers claimed, “media element”, to correspond to the cited, “patch”.).
Yuchai & Mao are combinable because they are from the same field of endeavor.
At the time before the effective filing date, it would have been obvious to a person of ordinary skill in the art to utilize this processing the patch components of Mao in Yuchai’s invention.
The suggestion/motivation for doing so would have been as indicated on paragraph 0075, “to generate a predicted object classification for the object of interest represented in the object patches.”.
Yuchai and Mao do not expressly recite, during an inference stage when a codec comprising the at least one NN compresses or decompresses one or more media elements and the apparatus comprises the codec or the codec comprises the apparatus.
Wang recites during an inference stage when a codec comprising the at least one NN compresses or decompresses one or more media elements and the apparatus comprises the codec or the codec comprises the apparatus. (Please note, page 5, 5th paragraph. As indicated each scene or segment of an original media file is compressed using one of a plurality of codecs. A particular codec best suited to compressing each scene is selected from a codec library. An Al system can be used to determine whether a codec exists in the library that has previously been found to optimally compress a scene. The Al system can be implemented as a neural network, which can be used to select codec settings, such as preprocessing/postprocessing filters.).
Yuchai, Mao & Wang are combinable because they are from the same field of endeavor.
At the time of the invention, it would have been obvious to a person of ordinary skill in the art to utilize this during an inference stage when a codec comprising the at least one NN compresses or decompresses one or more media elements of Wang in Yuchai & Mao’s invention.
The suggestion/motivation for doing so would have been as indicated on page 5, 5th paragraph, “for optimizing the compression of video and audio signals.”.
Yuchai, Mao and Wang do not expressly recite, multiply a signal at a decoder side.
Ge recites multiply a signal at a decoder side. (Please note, paragraph 0096. As indicated a receiver or decoder may multiply the received signal vector by the inverse or reciprocals of the scalar and shift vectors at each depolarization step.).
Yuchai, Mao, Wang & Ge are combinable because they are from the same field of endeavor.
At the time of the invention, it would have been obvious to a person of ordinary skill in the art to utilize this multiply a signal at a decoder side of Ge in Yuchai, Mao and Wang’s invention.
The suggestion/motivation for doing so would have been as indicated on paragraph 0096, “for a receiver or a decoder to recover the likelihood measurements of the target signals (usually with added noise) back to the likelihood measurements of the source distribution.”.
Therefore, it would have been obvious to combine Yuchai, Mao, Wang with Ge to obtain the invention as specified in claim 82.
Regarding claim 83, Ge recites, a feature map output by a convolutional layer, or a feature map output by a fully-connected layer. (Please note, paragraph 0083. As indicated for a convolutional code, because its decoder uses a convolutional decoder, the information bits have equal likelihood distribution. In this sense, both the input and output of a convolutional code have equal likelihood distributions.).
Regarding claim 89, Mao teaches, at least one of a decoder side neural network, a portion of the decoder side neural network, an encoder side neural network, or a portion of the encoder side neural network. (Please note, paragraph 0012. As indicated the object classifier neural network can include a plurality of channel encoders and a classification portion, each channel encoder configured to independently process a different one of the first neural network inputs to generate an alternative representation of the sensor measurements represented by the first neural network input, the classification portion configured to process the alternative representations from the plurality of channel encoders and the first of the plurality of feature vectors to generate the object classification.).
Regarding claim 90, Ge recites, an NN post-processing filter; an NN in-loop filter; a learned probability model that is used for lossless coding; a decoder NN for an end-to-end learned codec; a NN that performs intra-frame prediction; a NN that performs inter-frame prediction; or a NN that performs inverse transform. (Please note, paragraph 0028. As indicated a processor configured to implement a neural network, the neural network being trained to perform binary classification between two candidate signal distributions in a signal space and to output a cross entropy value between the two candidate signal distributions. The two candidate signal distributions are determined to be satisfactory when the cross entropy value is at a maximum.).
Regarding claims (92-93; 99); analysis similar to those presented for claims (82-83; 89-90), respectively, are applicable.
Claims 84-85 and 94-95 are rejected under 35 U.S.C. 103(a) as being unpatentable over Yuchai et al. (CN112163599A image classification method based on multi-scale and multi-level fusion), hereinafter, “Yuchai”, in view of Mao et al. (USPAP 2020/0202,145), hereinafter, “Mao”, and further in view of Wang et al. (EP 3278559 B1, TRAINING END-TO-END VIDEO PROCESSES), hereinafter, “Wang”, as applied to claims 81 and 91 above, as applied to claims 81 and 91 above, and further in view of Milanfar et al. (USPAP 2020/0186,836), hereinafter, “Milanfar”.
Regarding claim 84, Yuchai teaches, train or finetune one or more additional parameters of at least one neural network (NN) or a portion of the at least one NN, wherein the one or more additional parameters comprise one or more scaling parameters. (Please note, paragraph 0038. As indicated for training feature fusion classifiers, a transfer learning strategy can be used for parameter learning. That is, first, the neural network parameters are pre-trained on a large-scale existing dataset, and then the neural network is transferred to a specific application domain, where the neural network parameters are fine-tuned using a small amount of data from that domain.).
Yuchai does not expressly teach, encode, decode, or process one or more media elements based on the at least one neural network or the portion of the at least one NN comprising the trained or finetuned one or more additional parameters.
Mao teaches, encode, decode, or process one or more media elements based on the at least one neural network or the portion of the at least one NN comprising the trained or finetuned one or more additional parameters. (Please note, paragraph 0075. As indicated the object classifier neural network processes the patch components of the training example and the selected feature vector in accordance with current values of the parameters of the network. In this regard, examiner considers claimed, “media element”, to correspond to the cited, “patch”.).
Yuchai & Mao are combinable because they are from the same field of endeavor.
At the time before the effective filing date, it would have been obvious to a person of ordinary skill in the art to utilize this processing the patch components of Mao in Yuchai’s invention.
The suggestion/motivation for doing so would have been as indicated on paragraph 0075, “to generate a predicted object classification for the object of interest represented in the object patches.”.
Yuchai and Mao do not expressly recite, during an inference stage when a codec comprising the at least one NN compresses or decompresses one or more media elements and the apparatus comprises the codec or the codec comprises the apparatus.
Wang recites during an inference stage when a codec comprising the at least one NN compresses or decompresses one or more media elements and the apparatus comprises the codec or the codec comprises the apparatus. (Please note, page 5, 5th paragraph. As indicated each scene or segment of an original media file is compressed using one of a plurality of codecs. A particular codec best suited to compressing each scene is selected from a codec library. An Al system can be used to determine whether a codec exists in the library that has previously been found to optimally compress a scene. The Al system can be implemented as a neural network, which can be used to select codec settings, such as preprocessing/postprocessing filters.).
Yuchai, Mao & Wang are combinable because they are from the same field of endeavor.
At the time of the invention, it would have been obvious to a person of ordinary skill in the art to utilize this during an inference stage when a codec comprising the at least one NN compresses or decompresses one or more media elements of Wang in Yuchai & Mao’s invention.
The suggestion/motivation for doing so would have been as indicated on page 5, 5th paragraph, “for optimizing the compression of video and audio signals.”
Yuchai, Mao and Wang do not expressly recite, updating the one or more scaling parameters by using a combination operation to combine the one or more scaling parameters with associated updates.
Milanfar recites updating the one or more scaling parameters by using a combination operation to combine the one or more scaling parameters with associated updates. (Please note, paragraph 0057. As indicated while the method 300 illustrates the use of a single encoded image 310 to train a set of scaling factors 305, it is anticipated that a large set of encoded images (e.g., from an online database of such images) will be used to train the set of scaling factors 305. Accordingly, the method 300 may be performed on the plurality of encoded images to update the set of scaling factors serially (e.g., generating individual updates to the set of scaling factors based on the comparison data from a single encoded image) or in parallel (e.g., combining the comparison data from a set of encoded images to effect each individual update to the set of scaling factors).).
Yuchai, Mao, Wang & Milanfar are combinable because they are from the same field of endeavor.
At the time of the invention, it would have been obvious to a person of ordinary skill in the art to utilize this updating the one or more scaling parameters by using a combination operation to combine the one or more scaling parameters with associated updates of Milanfar in Yuchai, Mao and Wang’s invention.
The suggestion/motivation for doing so would have been as indicated on paragraph 0057, “to effect each individual update to the set of scaling factors).”.
Therefore, it would have been obvious to combine Yuchai, Mao, Wang with Milanfar to obtain the invention as specified in claim 84.
Regarding claim 85, Milanfar recites, wherein the combination operation comprises a summation operation, a multiplication operation, a predefined operation, or an operation selected from available operations. (Please note, paragraph 0050. As indicated a decoding software could multiply the image-content coefficients in the encoded image with corresponding elements of the quantization table 230 in order to “scale up” the quantized coefficients so that they may be transformed.).
Regarding claim 94-95 analysis similar to those presented for claims 84-85, respectively, are applicable.
Claim 101 is rejected under 35 U.S.C. 103(a) as being unpatentable over Yuchai et al. (CN112163599A image classification method based on multi-scale and multi-level fusion), hereinafter, “Yuchai”, in view of Mao et al. (USPAP 2020/0202,145), hereinafter, “Mao”, and further in view of Wang et al. (EP 3278559 B1, TRAINING END-TO-END VIDEO PROCESSES), hereinafter, “Wang”, and further in view of Milanfar et al. (USPAP 2020/0186,836), hereinafter, “Milanfar”, as applied to claims 81 and 91 above, and further in view of Liu et al. (USPAP 2020/0394,523), hereinafter, “Liu”.
Regarding claim 101:
Yuchai, Mao, Wang and Milanfar, do not expressly recite, wherein the associated updates are part of a weight update to the at least one NN.
Liu recites wherein the associated updates are part of a weight update to the at least one NN. (Please note, paragraph 0055. As indicated the corresponding gradient is subtracted from each weight in the neural network, then the weight is updated).
Yuchai, Mao, Wang, Milanfar and Liu are combinable because they are from the same field of endeavor.
At the time of the invention, it would have been obvious to a person of ordinary skill in the art to utilize this wherein the associated updates are part of a weight update to the at least one NN of Liu, Milanfar Yuchai, Mao and Wang’s invention.
The suggestion/motivation for doing so would have been as indicated on paragraph 0055, “to reduce errors.”.
Therefore, it would have been obvious to combine Milanfar, Yuchai, Mao, Wang with Liu to obtain the invention as specified in claim 101.
Allowable Subject Matter
Claims 86-88 and 96-98 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.
The following is a statement of reasons for the indication of allowable subject matter: The closest applied Prior Art of record fails to disclose or reasonably suggest wherein the apparatus is further caused to obtain one or more updates to the one or more scaling parameters, and wherein the one or more updates to the one or more scaling parameters are comprised in a syntax structure specifying a validity scope of the one or more updates.
Examiner’s Note
The examiner cites particular figures, paragraphs, columns and line numbers in the references as applied to the claims for the convenience of the applicant. Although the specified citations are representative of the teachings in the art and are applied to the specific limitations within the individual claims, other passages and figures may apply as well.
It is respectfully requested that, in preparing responses, the applicant fully consider the references in their entirety as potentially teaching all or part of the claimed invention, as well as the context of the passage as taught by the prior art or disclosed by the examiner.
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 AMIR ALAVI whose telephone number is (571)272-7386. The examiner can normally be reached on M-F from 8:00-4:30.
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, Vu Le can be reached at (571)272-7332. 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.
/AMIR ALAVI/Primary Examiner, Art Unit 2668 Tuesday, April 21, 2026
Prosecution Insights