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 .
Amendment
Applicant submitted amendments on 06/10/2026. The Examiner acknowledges the amendment and has reviewed the claims accordingly.
Priority
Receipt is acknowledged that application is a continuation application of PCT RU2021/000496. Priority to RU2021/000496 with a priority date of 11/11/2021 is acknowledged under 35 USC 119(e) and 37 CFR 1.78.
Information Disclosure Statement
The IDS dated 09/16/2024, 09/19/2024, and 02/09/2025 have been considered and placed in the application file.
Applicant Arguments:
In regards to the argument on Argument 1, Applicant/s state/s “Claim 17 has now been amended to correct these informalities. Reconsideration and withdrawal of the objection to claim 17 is therefore respectfully requested.” (See Remarks Pg 10, paragraph 6). Therefore the claim objection on Claim 17 should be withdrawn.
In regards to the argument on Argument 2, Applicant/s state/s “Claims 6-7, 15, and 17 have been amended to address the antecedent basis issues. In addition, claims 6 and 15 have been amended to specify that the sizes recited in these claims refer to the height and width dimensions. Claims 7 and 17 have been amended to specify that the sizes recited in these claims refer to the channel dimension.” (See Remarks Pg 10, paragraph 8). Therefore 35 U.S.C. § 112(b) rejection on Claims 6-7, 15, and 17 should be withdrawn.
In regards to the argument on Argument 3, Applicant/s state/s “It is respectfully submitted that, as amended, claims 1-6, 8, and 24 do not recite an abstract idea, are not directed to an abstract idea, and recite a number of particular limitations that, in combination, amount to significantly more than any judicial exception. Specifically, claims 1-6, 8, and 24 recite features, which alone and in combination, provide solutions to overcome particular problems specifically arising in the realm of image and video coding and, in particular, image and video coding comprising conditional image compression.” (See Remarks Pg 11, paragraph 2) and “as amended, recite a specific technological implementation that improves the technical field of image and video coding by enabling independent coding and decoding of image components, increasing processing parallelization, and reducing memory demands while maintaining reconstructed image quality.” (See Remarks Pg 13, paragraph 3). Therefore U.S.C 101 rejection on Claims 1-6, 8, and 24 should be withdrawn.
In regards to the argument on Argument 4, Applicant/s state/s “claim 1 has now been amended to recite that the first bitstream is generated based on a first entropy model and that the second bitstream is generated based on a second entropy model that is different from the first entropy model. Thus, the amended claim requires two distinct entropy models respectively used for generating the first and second bitstreams during the encoding process. Besenbruch does not disclose, the use of different entropy models for generating separate bitstreams as now claimed. Rather, the cited disclosures merely relate to entropy encoding and decoding generally and do not teach or suggest the specific configuration now recited in amended claim 1.” (See Remarks Pg 14, paragraph 4). Therefore U.S.C 102 and 103 rejections on Claims 1-8, and 24 should be withdrawn.
In regards to the argument on Argument 5, Applicant/s state/s “the cited disclosures of Besenbruch merely teach the use of different entropy models on the encoding side and the decoding side. In contrast claims 9, 18, and 25 require two distinct entropy models that are both used on the decoding side to reconstruct the primary and secondary components from the respective bitstreams. Accordingly, Besenbruch does not disclose or suggest the claimed use of first and second distinct entropy models in the decoding process.” (See Remarks Pg 15, paragraph 3). Therefore U.S.C 103 rejection on Claims 9-23 and 25 should be withdrawn.
Examiner’s Responses:
In response to Argument 1, Applicant’s arguments, see Remarks, filed 06/10/2026, with respect to the objection of Claim 17 have been considered and are persuasive. Therefore, the objection has been withdrawn due to amendments.
In response to Argument 2, Applicant’s arguments, see Remarks, filed 06/10/2026, with respect to the 35 U.S.C. § 112(b) rejection on Claims 6-7, 15, and 17 have been considered and are persuasive. Therefore, the35 U.S.C. § 112(b) rejection has been withdrawn due to amendments.
In response to Argument 3, Applicant’s arguments, see Remarks, filed 06/10/2026, with respect to the 35 U.S.C. § 101 rejection on Claims 1-6, 8, and 24 have been considered and are persuasive. Therefore, the 35 U.S.C. § 101 rejection on Claims 1-6, 8, and 24 has been withdrawn due to amendments.
In response to Argument 4, Applicant’s arguments, see Remarks, filed 06/10/2026 with respect to the U.S.C 103 rejections of Claims 1-8 and 24 have been considered but are moot in view of new ground(s) of rejection caused by the amendments. A new ground(s) of rejection is made for claims 1-8 and 24 under 35 U.S.C. 103 in view of Zhu (Zhu, Linwei, et al. "Deep learning-based chroma prediction for intra versatile video coding." IEEE Transactions on Circuits and Systems for Video Technology 31.8 (2020): 3168-3181.) in view of Besenbruch et al (WO Patent Publication WO2021220008 A1, hereafter referred to as Besenbruch) in further view of Van Rozendaal et al (US Patent Publication US 2022/0103839 A1, hereafter referred to as Van Rozendaal).
The Examiner finds that Besenbruch teaches on the amendment claim language “to generate a first bitstream” and “to generate a second bitstream” and “wherein the first bitstream is generated based on a first entropy model, and the second bitstream is generated based on a second entropy model” in the amended claims 1-8 and 24.
Specifically, Besenbruch, teaches using entropy encoding to create a first and second bitstream in ¶0119, ¶0183. Besenbruch also teaches that the first and second bitstream are generated based on entropy models in ¶0119, ¶0183, ¶0120. The Examiner also finds that Besenbruch does teach on the two different models wherein decoding is different from encoding indicating a separate model in ¶0120, and during prosecution, claims must be given their broadest reasonable interpretation while reading claim language in light of the specification as it would be interpreted by one of ordinary skill in the art. In re Am. Acad. of Sci. Tech. Ctr., 367 F.3d 1359, 1364 (Fed. Cir. 2004). In construing the meaning of claims terms, caution must be taken not to import limitations from the specification as “[i]t is the claims that measure the invention.” See SRI Int’l v. Matsushita Elec. Corp. of Am., 775 F.2d 1107, 1121 (Fed. Cir. 1985) (en banc) The Examiner interprets that under broadest reasonable interpretation “different” has
no special definition in the claims, and therefore can be interpreted as any type difference between the two models, including the encoding and decoding function as taught by Besenbruch.
However, for additional support an based on the change of scope necessitated by amendments an additional reference has been applied. Applicant argues that “claim 1 has now been amended to recite that the first bitstream is generated based on a first entropy model and that the second bitstream is generated based on a second entropy model that is different from the first entropy model. Thus, the amended claim requires two distinct entropy models respectively used for generating the first and second bitstreams during the encoding process. Besenbruch does not disclose, the use of different entropy models for generating separate bitstreams as now claimed. Rather, the cited disclosures merely relate to entropy encoding and decoding generally and do not teach or suggest the specific configuration now recited in amended claim 1.” The Examiner interprets that Besenbruch does teach the concept of the of generating two bitstreams from two entropy models while the Van Rozendaal reference teaches the concept of the two models being different, therefore, the additional details of the functions of the main concepts as stated above by the applicant in the amendments is taught by Van Rozendaal in the details of the rejection below. The Examiner will maintain prior art Zhu and Besenbruch and details of the rejection are below.
In response to Argument 5, Applicant’s arguments, see Remarks, filed 06/10/2026, with respect to the U.S.C 103 rejections of Claims 9-23 and 25 have been considered but are not persuasive.
The Examiner finds that Besenbruch teaches on the amendment claim language “processing a first bitstream” and “processing a second bitstream” and “different from the first bitstream based on a second entropy model different from the first entropy model” in the claims 9-23 and 25.
Specifically, Besenbruch, teaches using entropy encoding to create a first and second bitstream in ¶0119, ¶0183. Besenbruch also teaches that the first and second bitstream are generated based on entropy models in ¶0093, ¶0119, ¶0183, ¶0120. Besenbruch also teaches entropy decoding the bitstream, wherein decoding is different from encoding indicating a separate model in ¶0120. Applicant argues that “the cited disclosures of Besenbruch merely teach the use of different entropy models on the encoding side and the decoding side. In contrast claims 9, 18, and 25 require two distinct entropy models that are both used on the decoding side to reconstruct the primary and secondary components from the respective bitstreams. Accordingly, Besenbruch does not disclose or suggest the claimed use of first and second distinct entropy models in the decoding process”. The examiner respectfully disagrees, we determine claim scope not solely on the basis of claim language, but also on giving claims their broadest reasonable construction in light of the specification as it would be interpreted by one of ordinary skill in the art. In re Am. Acad. of Sci. Tech. Ctr., 367 F.3d 1359, 1364 (Fed. Cir. 2004). See also Superguide Corp. v. DirecTV Enterprises, Inc., 358 F.3d 870, 875 (Fed. Cir. 2004) (“Though understanding the claim language may be aided by explanations contained in the written description, it is important not to import into a claim limitations that are not part of the claim.”). The Examiner interprets that under broadest reasonable interpretation “different” has no special definition in the claims, and therefore can be interpreted as any type of difference between the two models, including the encoding and decoding function as taught by Besenbruch.
Additionally, in response to applicant's argument that the references fail to show certain features of applicant’s invention, it is noted that the features upon which applicant relies (i.e., two distinct entropy models that are both used on the decoding side) are not recited in the rejected claim(s). Although the claims are interpreted in light of the specification, limitations from the specification are not read into the claims. See In re Van Geuns, 988 F.2d 1181, 26 USPQ2d 1057 (Fed. Cir. 1993). Therefore, the Examiner interprets that Besenbruch does teach the concept of the of generating two bitstreams from two entropy models and that models are different. The Examiner will maintain prior art Zhu and Besenbruch and details of the rejection are below.
Claim Interpretation
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.
Under MPEP 2143.03, "All words in a claim must be considered in judging the patentability of that claim against the prior art." In re Wilson, 424 F.2d 1382, 1385, 165 USPQ 494, 496 (CCPA 1970). As a general matter, the grammar and ordinary meaning of terms as understood by one having ordinary skill in the art used in a claim will dictate whether, and to what extent, the language limits the claim scope. Language that suggests or makes a feature or step optional but does not require that feature or step does not limit the scope of a claim under the broadest reasonable claim interpretation. In addition, when a claim requires selection of an element from a list of alternatives, the prior art teaches the element if one of the alternatives is taught by the prior art. See, e.g., Fresenius USA, Inc. v. Baxter Int’l, Inc., 582 F.3d 1288, 1298, 92 USPQ2d 1163, 1171 (Fed. Cir. 2009).
Claim 6 and 15 recite “at least one of ” then listing “the height and width dimensions” Since “at least one of” is disjunctive, any one of the elements found in the prior art is sufficient to reject the claim. While citations have been provided for completeness and rapid prosecution, only one element is required. Because, on balance, it appears the disjunctive interpretation enjoys the most specification support and for that reason the disjunctive interpretation (one of A, B OR C) is being adopted for the purposes of this Office Action. Applicant’s comments and/or amendments relating to this issue are invited to clarify the claim language and the prosecution history.
Claim 7 recite “or ” then listing “wherein the size of the first latent tensor in the channel dimension is larger than, smaller than or equal to the size of the second latent tensor in the channel dimension, or wherein the first tensor is transformed into the first latent tensor through a first neural network, and the concatenated tensor is transformed into the second latent tensor through a second neural network different from the first neural network, or wherein the first bitstream is generated based on a first entropy model, and the second bitstream is generated based on a second entropy model different from the first entropy model.” Since “or” is disjunctive, any one of the elements found in the prior art is sufficient to reject the claim. While citations have been provided for completeness and rapid prosecution, only one element is required. Because, on balance, it appears the disjunctive interpretation enjoys the most specification support and for that reason the disjunctive interpretation (one of A, B OR C) is being adopted for the purposes of this Office Action. Applicant’s comments and/or amendments relating to this issue are invited to clarify the claim language and the prosecution history.
Claim 20 recite “or” then listing “wherein the primary component of the image is a luma component and the at least one secondary component of the image is a chroma component, or wherein the primary component of the image is a chroma component and the at least one secondary component of the image is a luma component.” Since “or” is disjunctive, any one of the elements found in the prior art is sufficient to reject the claim. While citations have been provided for completeness and rapid prosecution, only one element is required. Because, on balance, it appears the disjunctive interpretation enjoys the most specification support and for that reason the disjunctive interpretation (one of A, B OR C) is being adopted for the purposes of this Office Action. Applicant’s comments and/or amendments relating to this issue are invited to clarify the claim language and the prosecution history.
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.
Claims 1-8, and 24 are rejected under 35 U.S.C. 103 as unpatentable over Zhu (Zhu, Linwei, et al. "Deep learning-based chroma prediction for intra versatile video coding." IEEE Transactions on Circuits and Systems for Video Technology 31.8 (2020): 3168-3181.) in view of Besenbruch et al (WO Patent Publication WO2021220008 A1, hereafter referred to as Besenbruch) in further view of Van Rozendaal et al (US Patent Publication US 2022/0103839 A1, hereafter referred to as Van Rozendaal).
Regarding Claim 1, Zhu teaches a method of encoding (Zhu Pg 2 Col 2 ¶02 and Pg 5, Col 1, ¶05 and Pg 6 Col 1 ¶04 discloses employing an encoder-decoder structure and an encoder side of the neural network) at least a portion of an image (Zhu Pg 1 Col 1 ¶03, and Pg 2 Col 1 ¶01 discloses partitioning the image into blocks), the method applied to an electronic encoding device (Zhu Pg 8 Col1 ¶03 discloses using a Windows 10 Enterprise 64-bit operating system, to run the experiment) and comprising
encoding a primary component of the image (Zhu Fig 7a discloses the chroma encoding) independently from at least one secondary component of the image (Zhu Pg 2 Col 2 ¶01 and Pg 3 Col 2 ¶04 discloses two separate networks one to process luma and one to process chroma, meaning they are processed separately)
encoding the at least one secondary component of the image (Zhu Fig 5, Pg 7 Col 1 ¶1 discloses the luma down sampling network and encoding) using information from the primary component (Zhu Pg 3 Col 2 ¶04 discloses the model that performs the encoding using both luma and chroma pixels).
Zhu does not explicitly disclose to generate a first bitstream, to generate a second bitstream, wherein the first bitstream is generated based on a first entropy model, and the second
bitstream is generated based on a second entropy model.
Besenbruch is in the same field of image compression while preserving image quality. Further, Besenbruch teaches to generate a first bitstream (Besenbruch ¶0119 discloses entropy encoding the quantized latent into a bitstream) to generate a second bitstream (Besenbruch ¶0183 discloses entropy encoding the quantized z latent into a second bitstream),
wherein the first bitstream is generated based on a first entropy model (Besenbruch ¶0119 discloses entropy encoding the quantized latent into a bitstream) , and the second
bitstream is generated based on a second entropy model (Besenbruch ¶0183 discloses entropy encoding the quantized z latent into a second bitstream) ¶0120 discloses entropy decoding the bitstream, wherein decoding is different from encoding indicating a separate model).
Therefore, it would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Zhu by incorporating the transformation of the tensors into latent tensors and bitstreams in both actions of reducing dimensionality and increasing dimensionality as taught by Besenbruch; to make an invention that utilizes the reduced dimensionality to ease the computational burden; thus one of ordinary skilled in the art would be motivated to combine the references since there is a need to increase the compression, while preserving displayed image quality, or to increase the displayed image quality, while not increasing the amount of data that is actually transmitted across the communications networks. This would help to reduce the demands on communications networks, compared to the demands that otherwise would be made as disclosed by Besenbruch in ¶0004.
Thus, the claimed subject matter would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention.
Zhu and Besenbruch in combination do not explicitly disclose different from the first entropy model.
Van Rozendaal is in the same field of image compression while preserving image quality. Further, Van Rozendaal teaches different from the first entropy model (VAN ROZENDAAL ¶0127, ¶0144, ¶0147, ¶0156 and Fig 5A, Fig 7 discloses the arithmetic encoder used to entropy code to latent code can be different from the other encoder and the same concept applies to encoder -decoder and decoder -decoder models).
Therefore, it would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Zhu in view of Besenbruch by incorporating different entropy models as taught by Van Rozendaal; to make an invention that utilizes the reduced dimensionality to ease the computational burden and increase computation speed; thus one of ordinary skilled in the art would be motivated to combine the references since there is a need to compress video data into a form that uses a lower bit rate,
while avoiding or minimizing degradations in video quality. With ever-evolving video services becoming available and the increasing demands in large amounts of video data, coding techniques with better performance and efficiency are needed as disclosed by Van Rozendaal in ¶0003.
Thus, the claimed subject matter would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention.
Regarding Claim 2, Zhu in view of Besenbruch in further view of Van Rozendaal teaches the method according to claim 1, wherein the primary component (Zhu Fig 7a discloses the chroma encoding) and the at least one secondary component (Zhu Fig 5, Pg 7 Col 1 ¶1 discloses the luma down sampling network and encoding) are encoded concurrently (Zhu Pg 7 Col 1 ¶02, Pg 4 Col 1 ¶04 discloses the networks operating simultaneously). See Claim 1 for rationale, its parent claim.
Regarding Claim 3, Zhu in view of Besenbruch in further view of Van Rozendaal teaches the method according to claim 1, wherein the primary component of the image is a luma component (Zhu Fig 5, Pg 7 Col 1 ¶1 discloses the luma down sampling network and encoding) and the at least one secondary component of the image is a chroma component (Zhu Fig 7a discloses the chroma encoding). See Claim 1 for rationale, its parent claim.
Regarding Claim 4, Zhu in view of Besenbruch in further view of Van Rozendaal teaches the method according to claim 1, wherein the primary component of the image is a chroma component (Zhu Fig 7a discloses the chroma encoding) and the at least one secondary component of the image is a luma component (Zhu Fig 5, Pg 7 Col 1 ¶1 discloses the luma down sampling network and encoding). See Claim 1 for rationale, its parent claim.
Regarding Claim 5, Zhu in view of Besenbruch in further view of Van Rozendaal teaches the method according to claim 1, wherein the encoding the primary component (Zhu Fig 7a discloses the chroma encoding) to generate a first bitstream (Besenbruch ¶0119 discloses entropy encoding the quantized latent into a bitstream); further comprises:
representing the primary component by a first tensor (Zhu Fig 5 discloses the chroma predication network arranged into n x n sizes and Pg 7 Col 2 ¶02 discloses TensorFlow being used for network training which implies that the fundamental data structure of the networks are tensors);
transforming the first tensor into a first latent tensor (Besenbruch ¶0093 discloses the first computer system is configured to quantize the latent representation to produce a quantized latent) ; and
processing the first latent tensor to generate the first bitstream (Besenbruch ¶0094 discloses the first computer system is configured to entropy encode the quantized latent into a bitstream ¶0119 discloses entropy encoding the quantized latent into a bitstream); and
wherein the encoding the at least one secondary component (Zhu Fig 5, Pg 7 Col 1 ¶1 discloses the luma down sampling network and encoding) to generate a second bitstream (Besenbruch ¶0183 discloses entropy encoding the quantized z latent into a second bitstream), further comprises:
representing the at least one secondary component by a second tensor different from the first tensor (Zhu Fig 5 discloses the luma network arranged into n x n sizes and Pg 7 Col 2 ¶02 discloses TensorFlow being used for network training which implies that the fundamental data structure of the networks are tensors) ;
concatenating the second tensor and the first tensor to obtain a concatenated tensor (Zhu Fig 5 discloses concatenating the tensors to output a chroma predication network that utilizes tensors);
transforming the concatenated tensor into a second latent tensor (Besenbruch ¶0093 discloses the first computer system is configured to quantize the latent representation to produce a quantized latent); and
processing the second latent tensor to generate the second bitstream (Besenbruch ¶0094 discloses the first computer system is configured to entropy encode the quantized latent into a bitstream ¶0183 discloses entropy encoding the quantized z latent into a second bitstream)). See Claim 1 for rationale, its parent claim.
Regarding Claim 6, Zhu in view of Besenbruch in further view of Van Rozendaal teaches the method according to claim 1, wherein the encoding the primary component (Zhu Fig 7a discloses the chroma encoding) to generate the first bitstream (Besenbruch ¶0094 discloses the first computer system is configured to entropy encode the quantized latent into a bitstream ¶0119 discloses entropy encoding the quantized latent into a bitstream)further comprises:
representing the primary component by a first tensor (Zhu Fig 5 discloses the chroma predication network arranged into n x n sizes and Pg 7 Col 2 ¶02 discloses TensorFlow being used for network training which implies that the fundamental data structure of the networks are tensors) having a height dimension and a width dimension (Zhu Pg 4 Col 2 ¶04 discloses the component being the size of 4N x 4N which the examiner is interpreting as the height and width of the component);
transforming the first tensor into a first latent tensor (Besenbruch ¶0093 discloses the first computer system is configured to quantize the latent representation to produce a quantized latent) ; and
processing the first latent tensor to generate the first bitstream (Besenbruch ¶0094 discloses the first computer system is configured to entropy encode the quantized latent into a bitstream ¶0119 discloses entropy encoding the quantized latent into a bitstream); and
wherein the encoding the at least one secondary component (Zhu Fig 5, Pg 7 Col 1 ¶1 discloses the luma down sampling network and encoding) to generate a second bitstream (Besenbruch ¶0183 discloses entropy encoding the quantized z latent into a second bitstream) further comprises:
representing the at least one secondary component by a second tensor different from the first tensor (Zhu Fig 5 discloses the luma network arranged into n x n sizes and Pg 7 Col 2 ¶02 discloses TensorFlow being used for network training which implies that the fundamental data structure of the networks are tensors) and having a height dimension and a width dimension (Zhu Pg 4 Col 2 ¶04 discloses the component being the size of 4N x 4N which the examiner is interpreting as the height and width of the component);
determining whether a size or a sub-pixel offset of samples of the second tensor in at least one of the height and width dimensions differs from a size or sub-pixel offset of samples in at least one of the height and width dimensions of the first tensor (Zhu Pg 7 Col 2 ¶03 discloses the chroma and luma having different partition and block sizes), and based on a determination that the size or sub-pixel offset of samples of the second tensor in at least one of the height and width dimensions(Zhu Pg 4 Col 2 ¶04 discloses the component being the size of 4N x 4N which the examiner is interpreting as the height and width of the component) differs from the size or sub-pixel offset of samples of the first tensor in at least one of the height and width dimensions (Zhu Pg 4 Col 2 ¶04 discloses the component being the size of 4N x 4N which the examiner is interpreting as the height and width of the component), adjusting the sample locations of the first tensor to match the sample locations of the second tensor to obtain an adjusted first tensor (Zhu Pg 7 Col 2 ¶03- Pg 8 Col1 ¶01-¶02 discloses motion compensation methods that handle the differences in images of rotation, zooming and shearing when processing both components including shape adaptive transform);
concatenating the second tensor and the adjusted first tensor to obtain a concatenated tensor (Zhu Fig 5 discloses concatenating the tensors to output a chroma predication network that utilizes tensors) only based on a determination that the size or sub-pixel offset of samples of the second tensor in at least one of the height and width dimensions (Zhu Pg 4 Col 2 ¶04 discloses the component being the size of 4N x 4N which the examiner is interpreting as the height and width of the component), differs from the size or sub-pixel offset of samples of the first tensor (Zhu Pg 7 Col 2 ¶03 discloses the chroma and luma having different partition and block sizes), in at least one of the height and width dimensions (Zhu Pg 4 Col 2 ¶04 discloses the component being the size of 4N x 4N which the examiner is interpreting as the height and width of the component), and else concatenating the second tensor and the first tensor to obtain a concatenated tensor(Zhu Fig 5 discloses concatenating the tensors to output a chroma predication network that utilizes tensors);
transforming the concatenated tensor into a second latent tensor (Besenbruch ¶0093 discloses the first computer system is configured to quantize the latent representation to produce a quantized latent); and
processing the second latent tensor to generate the second bitstream (Besenbruch ¶0094 discloses the first computer system is configured to entropy encode the quantized latent into a bitstream). See Claim 1 for rationale, its parent claim.
Regarding Claim 7, Zhu in view of Besenbruch in further view of Van Rozendaal teaches the method according to claim 6, wherein the first latent tensor comprises a channel dimension (Besenbruch Pg 103 ¶02 discloses where the tensor to be modelled is split into blocks (in spatial and or channel dimensions) and the second latent tensor comprises a channel dimension (Besenbruch Pg 103 ¶02 discloses where the tensor to be modelled is split into blocks (in spatial and or channel dimensions) and
wherein a channel dimension size (Besenbruch Pg 96 ¶02 discloses changing the channel dimensionality size if needed) of the first latent tensor is larger than, smaller than or equal to the a channel dimension size (Besenbruch Pg 96 ¶02 discloses changing the channel dimensionality size if needed) of the second latent tensor (Besenbruch Pg 103 ¶02 discloses that the channel dimension size can differ and can be adjusted to match the input) , or
wherein the first tensor is transformed into the first latent tensor through a first neural network (Besenbruch ¶0104 discloses using a first neural network to produce a latent representation), and the concatenated tensor (Zhu Fig 5 discloses concatenating the tensors to output a chroma predication network that utilizes tensors) is transformed into the second latent tensor through a second neural network different from the first neural network (Besenbruch ¶0105-¶0106 discloses using a second neural network to quantize the latent tensor). See Claim 1 for rationale, its parent claim.
Regarding Claim 8, Zhu teaches a method of encoding (Zhu Pg 2 Col 2 ¶02 and Pg 5, Col 1, ¶05 and Pg 6 Col 1 ¶04 discloses employing an encoder-decoder structure and an encoder side of the neural network) at least a portion of an image (Zhu Pg 1 Col 1 ¶03, and Pg 2 Col 1 ¶01 discloses partitioning the image into blocks), the method applied to an electronic encoding device (Zhu Pg 8 Col1 ¶03 discloses using a Windows 10 Enterprise 64-bit operating system, to run the experiment) and comprising:
providing a residual (Zhu Pg 3 Col 2 ¶02 discloses a residual including chroma residual signal and luma residual signal) comprising a primary residual component for a primary component of the image (Zhu Pg 3 Col 2 ¶02 discloses a residual including chroma residual signal) and at least one secondary residual component for at least one secondary component (Zhu Pg 3 Col 2 ¶02 discloses a residual including chroma residual signal and luma residual signal) of the image that is different from the primary component (Zhu Pg 3 Col 2 ¶02 discloses a residual including chroma residual signal and luma residual signal which are different);
encoding (Zhu Fig 7a discloses the chroma encoding) the primary residual component (Zhu Pg 3 Col 2 ¶02 discloses a residual including chroma residual signal and luma residual signal) independently from the at least one secondary residual component (Zhu Pg 2 Col 2 ¶01 and Pg 3 Col 2 ¶04 discloses two separate networks one to process luma and one to process chroma, meaning they are processed separately); and
encoding (Zhu Fig 5, Pg 7 Col 1 ¶1 discloses the luma down sampling network and encoding) the at least one secondary residual component (Zhu Pg 3 Col 2 ¶02 discloses a residual including chroma residual signal and luma residual signal) using information from the primary residual component (Zhu Pg 3 Col 2 ¶04 discloses the model that performs the encoding using both luma and chroma pixels).
Zhu does not explicitly disclose to generate a first bitstream to generate a second bitstream wherein the first bitstream is generated based on a first entropy model, and the second bitstream is generated based on a second entropy model.
Besenbruch is in the same field of image compression while preserving image quality. Further, Besenbruch teaches to generate a first bitstream (Besenbruch ¶0119 discloses entropy encoding the quantized latent into a bitstream)
to generate a second bitstream (Besenbruch ¶0183 discloses entropy encoding the quantized z latent into a second bitstream),
wherein the first bitstream is generated based on a first entropy model (Besenbruch ¶0119 discloses entropy encoding the quantized latent into a bitstream) , and the second
bitstream is generated based on a second entropy model (Besenbruch ¶0183 discloses entropy encoding the quantized z latent into a second bitstream) ¶0120 discloses entropy decoding the bitstream, wherein decoding is different from encoding indicating a separate model).
Therefore, it would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Zhu by incorporating the transformation of the tensors into latent tensors and bitstreams in both actions of reducing dimensionality and increasing dimensionality as taught by Besenbruch; to make an invention that utilizes the reduced dimensionality to ease the computational burden; thus one of ordinary skilled in the art would be motivated to combine the references since there is a need to increase the compression, while preserving displayed image quality, or to increase the displayed image quality, while not increasing the amount of data that is actually transmitted across the communications networks. This would help to reduce the demands on communications networks, compared to the demands that otherwise would be made as disclosed by Besenbruch in ¶0004.
Thus, the claimed subject matter would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention.
Zhu and Besenbruch in combination do not explicitly disclose different from the first entropy model.
Van Rozendaal is in the same field of image compression while preserving image quality. Further, Van Rozendaal teaches different from the first entropy model (VAN ROZENDAAL ¶0127, ¶0144, ¶0147, ¶0156 and Fig 5A, Fig 7 discloses the arithmetic encoder used to entropy code to latent code can be different from the other encoder and the same concept applies to encoder -decoder and decoder -decoder models).
Therefore, it would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Zhu in view of Besenbruch by incorporating different entropy models as taught by Van Rozendaal; to make an invention that utilizes the reduced dimensionality to ease the computational burden and increase computation speed; thus one of ordinary skilled in the art would be motivated to combine the references since there is a need to compress video data into a form that uses a lower bit rate,
while avoiding or minimizing degradations in video quality. With ever-evolving video services becoming available and the increasing demands in large amounts of video data, coding techniques with better performance and efficiency are needed as disclosed by Van Rozendaal in ¶0003.
Thus, the claimed subject matter would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention.
Regarding Claim 24, Zhu teaches a processing apparatus (Zhu Pg 8 Col1 ¶03 discloses using a Windows 10 Enterprise 64-bit operating system, to run the experiment) for encoding at least a portion of an image (Zhu Pg 2 Col 2 ¶02 and Pg 5, Col 1, ¶05 and Pg 6 Col 1 ¶04 discloses employing an encoder-decoder structure and an encoder side of the neural network Pg 1 Col 1 ¶03, and Pg 2 Col 1 ¶01 discloses partitioning the image into blocks), the processing apparatus comprising a processing circuitry (Zhu Pg 8 Col1 ¶03 discloses using a Windows 10 Enterprise 64-bit operating system including a CPU, to run the experiment) configured for:
encoding a primary component of the image (Zhu Fig 7a discloses the chroma encoding) independently from at least one secondary component of the image (Zhu Pg 2 Col 2 ¶01 and Pg 3 Col 2 ¶04 discloses two separate networks one to process luma and one to process chroma, meaning they are processed separately); and
encoding the at least one secondary component of the image (Zhu Fig 5, Pg 7 Col 1 ¶1 discloses the luma down sampling network and encoding) using information from the primary component (Zhu Pg 3 Col 2 ¶04 discloses the model that performs the encoding using both luma and chroma pixels).
Zhu does not explicitly disclose to generate a first bitstream to generate a second bitstream wherein the first bitstream is generated based on a first entropy model and the second
bitstream is generated based on a second entropy model.
Besenbruch is in the same field of image compression while preserving image quality. Further, Besenbruch teaches to generate a first bitstream (Besenbruch ¶0119 discloses entropy encoding the quantized latent into a bitstream);
to generate a second bitstream (Besenbruch ¶0183 discloses entropy encoding the quantized z latent into a second bitstream),
wherein the first bitstream is generated based on a first entropy model (Besenbruch ¶0119 discloses entropy encoding the quantized latent into a bitstream) , and the second
bitstream is generated based on a second entropy model (Besenbruch ¶0183 discloses entropy encoding the quantized z latent into a second bitstream) ¶0120 discloses entropy decoding the bitstream, wherein decoding is different from encoding indicating a separate model).
Therefore, it would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Zhu by incorporating the transformation of the tensors into latent tensors and bitstreams in both actions of reducing dimensionality and increasing dimensionality as taught by Besenbruch; to make an invention that utilizes the reduced dimensionality to ease the computational burden; thus one of ordinary skilled in the art would be motivated to combine the references since there is a need to increase the compression, while preserving displayed image quality, or to increase the displayed image quality, while not increasing the amount of data that is actually transmitted across the communications networks. This would help to reduce the demands on communications networks, compared to the demands that otherwise would be made as disclosed by Besenbruch in ¶0004.
Thus, the claimed subject matter would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention.
Zhu and Besenbruch in combination do not explicitly disclose different from the first entropy model.
Van Rozendaal is in the same field of image compression while preserving image quality. Further, Van Rozendaal teaches different from the first entropy model (VAN ROZENDAAL ¶0127, ¶0144, ¶0147, ¶0156 and Fig 5A, Fig 7 discloses the arithmetic encoder used to entropy code to latent code can be different from the other encoder and the same concept applies to encoder -decoder and decoder -decoder models).
Therefore, it would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Zhu in view of Besenbruch by incorporating different entropy models as taught by Van Rozendaal; to make an invention that utilizes the reduced dimensionality to ease the computational burden and increase computation speed; thus one of ordinary skilled in the art would be motivated to combine the references since there is a need to compress video data into a form that uses a lower bit rate,
while avoiding or minimizing degradations in video quality. With ever-evolving video services becoming available and the increasing demands in large amounts of video data, coding techniques with better performance and efficiency are needed as disclosed by Van Rozendaal in ¶0003.
Thus, the claimed subject matter would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention.
Claims 9-23 and 25 are rejected under 35 U.S.C. 103 as unpatentable over Zhu (Zhu, Linwei, et al. "Deep learning-based chroma prediction for intra versatile video coding." IEEE Transactions on Circuits and Systems for Video Technology 31.8 (2020): 3168-3181.) in view of Besenbruch et al (WO Patent Publication WO2021220008 A1, hereafter referred to as Besenbruch).
Regarding Claim 9, Zhu teaches a method of reconstructing (Zhu Pg 3 Col 1 ¶04 and Pg 4 Col 2 ¶01-¶02 discloses reconstructing the luma and chroma pixels of the image) at least a portion of an image (Zhu Pg 1 Col 1 ¶03, and Pg 2 Col 1 ¶01 discloses partitioning the image into blocks), representing a primary component of the image (Zhu Fig 7a discloses the chroma encoding);
representing at least one secondary component (Zhu Fig 5, Pg 7 Col 1 ¶1 discloses the luma down sampling network and encoding) of the image using information from the first latent tensor (Zhu Pg 3 Col 2 ¶04 discloses the model that performs the encoding using both luma and chroma pixels).
Zhu does not explicitly disclose processing a first bitstream based on a first entropy model to obtain a first latent tensor; processing the first latent tensor to obtain a first tensor,
processing a second bitstream different from the first bitstream based on a second entropy model different from the first entropy model to obtain a second latent tensor different from the first latent tensor; and
processing the second latent tensor to obtain a second tensor.
Besenbruch is in the same field of image compression while preserving image quality. Further, Besenbruch teaches processing a first bitstream based on a first entropy model to obtain a first latent tensor (Besenbruch ¶0119 discloses entropy encoding the quantized latent into a bitstream); processing the first latent tensor to obtain a first tensor (Besenbruch ¶0093 discloses the first computer system is configured to quantize the latent representation to produce a quantized latent)
processing a second bitstream (Besenbruch ¶0094 discloses the first computer system is configured to entropy encode the quantized latent into a bitstream) different from the first bitstream based on a second entropy model different from the first entropy model (Besenbruch ¶0120 discloses entropy decoding the bitstream, wherein decoding is different from encoding indicating a separate model) to obtain a second latent tensor different from the first latent tensor (Besenbruch ¶0105-¶0106 discloses using a second neural network to quantize the latent tensor); and
processing the second latent tensor to obtain a second tensor (Besenbruch ¶0094 discloses the first computer system is configured to entropy encode the quantized latent into a bitstream).
Therefore, it would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Zhu by incorporating the transformation of the tensors into latent tensors and bitstreams in both actions of reducing dimensionality and increasing dimensionality as taught by Besenbruch; to make an invention that utilizes the reduced dimensionality to ease the computational burden; thus one of ordinary skilled in the art would be motivated to combine the references since there is a need to increase the compression, while preserving displayed image quality, or to increase the displayed image quality, while not increasing the amount of data that is actually transmitted across the communications networks. This would help to reduce the demands on communications networks, compared to the demands that otherwise would be made as disclosed by Besenbruch in ¶0004.
Thus, the claimed subject matter would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention.
Regarding Claim 10, Zhu in view of Besenbruch teaches the method according to claim 9, wherein the first latent tensor (Besenbruch ¶0093 discloses the first computer system is configured to quantize the latent representation to produce a quantized latent) is processed independently (Zhu Pg 2 Col 2 ¶01 and Pg 3 Col 2 ¶04 discloses two separate networks one to process luma and one to process chroma, meaning they are processed separately) from the processing of the second latent tensor (Besenbruch ¶0093 discloses the first computer system is configured to quantize the latent representation to produce a quantized latent). See Claim 9 for rationale, its parent claim.
Regarding Claim 11, Zhu in view of Besenbruch teaches the method according to claim 9, wherein the primary component of the image is a luma component (Zhu Fig 5, Pg 7 Col 1 ¶1 discloses the luma down sampling network and encoding) and the at least one secondary component of the image is a chroma component (Zhu Fig 7a discloses the chroma encoding). See Claim 9 for rationale, its parent claim.
Regarding Claim 12, Zhu in view of Besenbruch teaches the method according to claim 9, wherein the primary component of the image is a chroma component (Zhu Fig 7a discloses the chroma encoding) and the at least one secondary component of the image is a luma component (Zhu Fig 5, Pg 7 Col 1 ¶1 discloses the luma down sampling network and encoding). See Claim 9 for rationale, its parent claim.
Regarding Claim 13, Zhu in view of Besenbruch teaches the method according to claim 11, wherein the second tensor(Zhu Pg 2 Col 2 ¶01 and Pg 3 Col 2 ¶04 discloses two separate networks one to process luma and one to process chroma, meaning they are processed separately) represents two secondary components one of which being a chroma component and the other one being another chroma component (Zhu Pg 7 Col 1 ¶2 and Pg 7 Col 2 ¶03 discloses two chroma components being predicted). See Claim 9 for rationale, its parent claim.
Regarding Claim 14, Zhu in view of Besenbruch teaches the method according to claim 9, wherein
the processing of the first latent tensor (Besenbruch ¶0093 discloses the first computer system is configured to quantize the latent representation to produce a quantized latent) comprises transforming the first latent tensor into the first tensor (Besenbruch Pg 135 Section 12.2.5 discloses using decomposition which involves breaking down the tensor into lower dimensions); and
the processing of the second latent tensor (Besenbruch ¶0093 discloses the first computer system is configured to quantize the latent representation to produce a quantized latent) comprises concatenating the second latent tensor and the first latent tensor to obtain a concatenated tensor (Zhu Fig 5 discloses concatenating the tensors to output a chroma predication network that utilizes tensors) and transforming the concatenated tensor into the second tensor ( Besenbruch Pg 135 Section 12.2.4 discloses concatenating the blocks in a separate dimension to reduce the number of channels by half, creating a new tensor). See Claim 9 for rationale, its parent claim.
Regarding Claim 15, Zhu in view of Besenbruch teaches the method according to claim 9, wherein each of the first and second latent tensors has a height and a width dimension (Zhu Pg 4 Col 2 ¶04 discloses the component being the size of 4N x 4N which the examiner is interpreting as the height and width of the component) and
the processing of the first latent tensor (Besenbruch ¶0093 discloses the first computer system is configured to quantize the latent representation to produce a quantized latent) comprises transforming the first latent tensor into the first tensor (Besenbruch Pg 135 Section 12.2.5 discloses using decomposition which involves breaking down the tensor into lower dimensions); and
the processing of the second latent tensor (Besenbruch ¶0093 discloses the first computer system is configured to quantize the latent representation to produce a quantized latent) comprises determining whether a size or a sub-pixel offset of samples of the second latent tensor in at least one of the height and width dimensions differs from a size or sub-pixel offset of samples in at least one of the height and width dimensions of the first latent tensor (Zhu Pg 7 Col 2 ¶03 discloses the chroma and luma having different partition and block sizes), and when it is determined that the size or sub-pixel offset of samples of the second latent tensor
in at least one of the height and width Dimensions(Zhu Pg 4 Col 2 ¶04 discloses the component being the size of 4N x 4N which the examiner is interpreting as the height and width of the component), differs from the size or sub-pixel offset of samples of the first latent tensor, adjusting the sample locations of the first latent tensor in at least one of the height and width Dimensions(Zhu Pg 4 Col 2 ¶04 discloses the component being the size of 4N x 4N which the examiner is interpreting as the height and width of the component), to match the sample locations of the second latent tensor thereby obtaining an adjusted first latent tensor (Zhu Pg 7 Col 2 ¶03- Pg 8 Col1 ¶01-¶02 discloses motion compensation methods that handle the differences in images of rotation, zooming and shearing when processing both components including shape adaptive transform);
concatenating the second latent tensor and the adjusted first latent tensor to obtain a concatenated latent tensor (Zhu Fig 5 discloses concatenating the tensors to output a chroma predication network that utilizes tensors) only when it is determined that the size or sub-pixel offset of samples of the second latent tensor in at least one of the height and width Dimensions (Zhu Pg 4 Col 2 ¶04 discloses the component being the size of 4N x 4N which the examiner is interpreting as the height and width of the component), differs from the size or sub-pixel offset of samples of the first latent tensor (Zhu Pg 7 Col 2 ¶03 discloses the chroma and luma having different partition and block sizes) in at least one of the height and width Dimensions (Zhu Pg 4 Col 2 ¶04 discloses the component being the size of 4N x 4N which the examiner is interpreting as the height and width of the component) and else concatenating the second latent tensor and the first latent tensor to obtain a concatenated latent tensor (Zhu Fig 5 discloses concatenating the tensors to output a chroma predication network that utilizes tensors); and
transforming the concatenated latent tensor into the second tensor (Besenbruch Pg 135 Section 12.2.4 discloses concatenating the blocks in a separate dimension to reduce the number of channels by half, creating a new tensor). See rationale for Claim 9, its parent claim.
Regarding Claim 16, Zhu in view of Besenbruch teaches the method according to claim 9, wherein the first bitstream is processed by a first neural network (Besenbruch ¶0185-¶0189 discloses a computer that processes the second bitstream, the computer having two different neural networks) and the second bitstream is processed by a second neural network different from the first neural network (Besenbruch ¶0177-¶0183 discloses a computer that processes the second bitstream, the computer having two different neural networks).See Claim 9 for rationale, its parent claim.
Regarding Claim 17, Zhu in view of Besenbruch teaches the method according to claim 9, wherein the first latent tensor comprises a channel dimension (Besenbruch Pg 103 ¶02 discloses where the tensor to be modelled is split into blocks (in spatial and or channel dimensions) and the second latent tensor comprises a channel dimension (Besenbruch Pg 103 ¶02 discloses where the tensor to be modelled is split into blocks (in spatial and or channel dimensions) and wherein a channel dimension size (Besenbruch Pg 96 ¶02 discloses changing the channel dimensionality size if needed) of the first latent tensor is one of larger than, smaller than or equal to a channel dimension size (Besenbruch Pg 96 ¶02 discloses changing the channel dimensionality size if needed) of the second latent tensor (Besenbruch Pg 103 ¶02 discloses that the channel dimension size can differ and can be adjusted to match the input). See Claim 9 for rationale, its parent claim.
Regarding Claim 18, Zhu teaches a method of reconstructing (Zhu Pg 3 Col 1 ¶04 and Pg 4 Col 2 ¶01-¶02 discloses reconstructing the luma and chroma pixels of the image) at least a portion of an image (Zhu Pg 1 Col 1 ¶03, and Pg 2 Col 1 ¶01 discloses partitioning the image into blocks),
representing a primary residual component of a residual for a primary component of the image (Zhu Pg 3 Col 2 ¶02 discloses a residual including chroma residual signal),
representing at least one secondary residual component of the residual for at least one secondary component (Zhu Pg 3 Col 2 ¶02 discloses a residual including chroma residual signal and luma residual signal) of the image using information from the first latent tensor (Zhu Pg 3 Col 2 ¶04 discloses the model that performs the encoding using both luma and chroma pixels).
Zhu does not explicitly disclose processing a first bitstream based on a first entropy model to obtain a first latent tensor;
processing the first latent tensor to obtain a first tensor,
processing a second bitstream different from the first bitstream based on a second entropy model different from the first entropy model to obtain a second latent tensor different from the first latent tensor; and
processing the second latent tensor to obtain a second tensor.
Besenbruch is in the same field of image compression while preserving image quality. Further, Besenbruch teaches processing a first bitstream based on a first entropy model to obtain a first latent tensor (Besenbruch ¶0119 discloses entropy encoding the quantized latent into a bitstream); processing the first latent tensor to obtain a first tensor (Besenbruch ¶0093 discloses the first computer system is configured to quantize the latent representation to produce a quantized latent)
processing a second bitstream (Besenbruch ¶0094 discloses the first computer system is configured to entropy encode the quantized latent into a bitstream) different from the first bitstream based on a second entropy model different from the first entropy model (Besenbruch ¶0120 discloses entropy decoding the bitstream, wherein decoding is different from encoding indicating a separate model) to obtain a second latent tensor different from the first latent tensor (Besenbruch ¶0105-¶0106 discloses using a second neural network to quantize the latent tensor); and
processing the second latent tensor to obtain a second tensor (Besenbruch ¶0094 discloses the first computer system is configured to entropy encode the quantized latent into a bitstream).
Therefore, it would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Zhu by incorporating the transformation of the tensors into latent tensors and bitstreams in both actions of reducing dimensionality and increasing dimensionality as taught by Besenbruch; to make an invention that utilizes the reduced dimensionality to ease the computational burden; thus one of ordinary skilled in the art would be motivated to combine the references since there is a need to increase the compression, while preserving displayed image quality, or to increase the displayed image quality, while not increasing the amount of data that is actually transmitted across the communications networks. This would help to reduce the demands on communications networks, compared to the demands that otherwise would be made as disclosed by Besenbruch in ¶0004.
Thus, the claimed subject matter would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention.
Regarding Claim 19, Zhu in view of Besenbruch teaches the method according to claim 18, wherein the first latent tensor (Besenbruch ¶0093 discloses the first computer system is configured to quantize the latent representation to produce a quantized latent) is processed independently (Zhu Pg 2 Col 2 ¶01 and Pg 3 Col 2 ¶04 discloses two separate networks one to process luma and one to process chroma, meaning they are processed separately) from the processing of the second latent tensor (Besenbruch ¶0093 discloses the first computer system is configured to quantize the latent representation to produce a quantized latent). See Claim 18 for rationale, its parent claim.
Regarding Claim 20, Zhu in view of Besenbruch teaches the method according to claim 18, wherein the primary component of the image is a luma component (Zhu Fig 5, Pg 7 Col 1 ¶1 discloses the luma down sampling network and encoding) and the at least one secondary component of the image is a chroma component (Zhu Fig 7a discloses the chroma encoding), or
wherein the primary component of the image is a chroma component (Zhu Fig 7a discloses the chroma encoding) and the at least one secondary component of the image is a luma component (Zhu Fig 5, Pg 7 Col 1 ¶1 discloses the luma down sampling network and encoding). See Claim 18 for rationale, its parent claim.
Regarding Claim 21, Zhu in view of Besenbruch teaches the method according to claim 18, wherein
the processing of the first latent tensor (Besenbruch ¶0093 discloses the first computer system is configured to quantize the latent representation to produce a quantized latent) comprises transforming the first latent tensor into the first tensor (Besenbruch Pg 135 Section 12.2.5 discloses using decomposition which involves breaking down the tensor into lower dimensions); and
the processing of the second latent tensor comprises concatenating the second latent tensor (Zhu Fig 5 discloses concatenating the tensors to output a chroma predication network that utilizes tensors) and transforming the concatenated latent tensor into the second tensor ( Besenbruch Pg 135 Section 12.2.4 discloses concatenating the blocks in a separate dimension to reduce the number of channels by half, creating a new tensor). See Claim 18 for rationale, its parent claim.
Regarding Claim 22, Zhu in view of Besenbruch teaches the method according to claim 18, wherein the first bitstream is processed by a first neural network (Besenbruch ¶0185-¶0189 discloses a computer that processes the second bitstream, the computer having two different neural networks) and the second bitstream is processed by a second neural network different from the first neural network (Besenbruch ¶0177-¶0183 discloses a computer that processes the second bitstream, the computer having two different neural networks). See Claim 18 for rationale, its parent claim.
Regarding Claim 23, Zhu in view of Besenbruch teaches a non-transitory computer-readable medium comprising a code which when executed by one or more processors performs the method according to claim 18 (Besenbruch ¶0165, ¶0176, ¶0291, ¶0308 discloses a computer program product that is executed by processors). See Claim 18 for rationale, its parent claim.
Regarding Claim 25, Zhu teaches a processing apparatus (Zhu Pg 8 Col1 ¶03 discloses using a Windows 10 Enterprise 64-bit operating system, to run the experiment) for reconstructing (Zhu Pg 3 Col 1 ¶04 and Pg 4 Col 2 ¶01-¶02 discloses reconstructing the luma and chroma pixels of the image) at least a portion of an image, (Zhu Pg 2 Col 2 ¶02 and Pg 5, Col 1, ¶05 and Pg 6 Col 1 ¶04 discloses employing an encoder-decoder structure and an encoder side of the neural network Pg 1 Col 1 ¶03, and Pg 2 Col 1 ¶01 discloses partitioning the image into blocks), the processing apparatus comprising a processing circuitry (Zhu Pg 8 Col1 ¶03 discloses using a Windows 10 Enterprise 64-bit operating system including a CPU, to run the experiment);
representing a primary component of the image (Zhu Fig 7a discloses the chroma encoding);
representing at least one secondary component (Zhu Fig 5, Pg 7 Col 1 ¶1 discloses the luma down sampling network and encoding) of the image using information from the first latent tensor (Zhu Pg 3 Col 2 ¶04 discloses the model that performs the encoding using both luma and chroma pixels).
Zhu does not explicitly disclose processing a first bitstream based on a first entropy model to obtain a first latent tensor;
processing the first latent tensor to obtain a first tensor;
processing a second bitstream different from the first bitstream based on a second entropy model different from the first entropy model to obtain a second latent tensor different from the first latent tensor; and
processing the second latent tensor to obtain a second tensor.
Besenbruch is in the same field of image compression while preserving image quality. Further, Besenbruch teaches processing a first bitstream based on a first entropy model to obtain a first latent tensor (Besenbruch ¶0119 discloses entropy encoding the quantized latent into a bitstream);
processing the first latent tensor to obtain a first tensor (Besenbruch ¶0093 discloses the first computer system is configured to quantize the latent representation to produce a quantized latent)
processing a second bitstream (Besenbruch ¶0094 discloses the first computer system is configured to entropy encode the quantized latent into a bitstream) different from the first bitstream based on a second entropy model different from the first entropy model (Besenbruch ¶0120 discloses entropy decoding the bitstream, wherein decoding is different from encoding indicating a separate model) to obtain a second latent tensor different from the first latent tensor (Besenbruch ¶0105-¶0106 discloses using a second neural network to quantize the latent tensor); and
processing the second latent tensor to obtain a second tensor (Besenbruch ¶0094 discloses the first computer system is configured to entropy encode the quantized latent into a bitstream).
Therefore, it would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Zhu by incorporating the transformation of the tensors into latent tensors and bitstreams in both actions of reducing dimensionality and increasing dimensionality as taught by Besenbruch; to make an invention that utilizes the reduced dimensionality to ease the computational burden; thus one of ordinary skilled in the art would be motivated to combine the references since there is a need to increase the compression, while preserving displayed image quality, or to increase the displayed image quality, while not increasing the amount of data that is actually transmitted across the communications networks. This would help to reduce the demands on communications networks, compared to the demands that otherwise would be made as disclosed by Besenbruch in ¶0004.
Thus, the claimed subject matter would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention.
Reference Cited
The prior art made of record and not relied upon is considered pertinent to applicant’s disclosure.
US Patent US-20220191553-A1 to Auyeung et al. discloses method and apparatus for video coding that incorporates neural networks into the video coding process.
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 RACHEL ROBERTS whose telephone number is (571)272-6413. The examiner can normally be reached Monday- Friday 7:30am- 5:00pm.
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, Oneal Mistry can be reached on (313) 446-4912. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of 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.
/RACHEL L ROBERTS/Examiner, Art Unit 2674
/ONEAL R MISTRY/Supervisory Patent Examiner, Art Unit 2674
Prosecution Insights