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 .
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim(s) 1-3, 5-7, 10-12, 14-18 and 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over US 2023/0062509 A1 (“Blanch”) in view of A Hybrid Neural Network for Chroma Intra Prediction (“Li”) in further view of US 2022/0248025 A1 (“Deng”) (Note: Li was attached in the previous office action).
Regarding claim 1, Blanch discloses a method comprising:
determining one or more candidate models from a first type of models and a second type of models (e.g. see traditional angular modes, LM models or the disclosed neural network mode, e.g. see at least paragraphs [0016], [0021], [0072]; LM models are linear type and neural network mode is non-linear type);
selecting a chroma prediction model from the one or more candidate models (e.g. see information necessary to compute the prediction for the current block in a received bit stream, e.g. see at least paragraph [0068]; for example signaling the use and the determined intra-prediction mode, e.g. see at least paragraph [0070]; this can include receiving (or by extension deriving at the decoder) the chosen mode between traditional angular modes, LM models or the disclosed neural network mode, e.g. see at least paragraphs [0016], [0021], [0072]);
generating a prediction of the chroma block (e.g. see prediction signal, e.g. see at least paragraph [0070], for chroma prediction, e.g. see at least paragraphs [0021], [0072]) based on reference signals of the chroma block (e.g. see reference neighbouring samples, e.g. see at least paragraphs [0020]-[0021], [0076]) and the chroma prediction model (e.g. see traditional angular modes, LM models or the disclosed neural network mode, e.g. see at least paragraphs [0016], [0021], [0072]); and
determining a reconstruction of the chroma block based on the prediction of the chroma block and a residual of the chroma block (e.g. see reconstruction of the original picture block derived from residual signal and prediction block, e.g. see at least paragraph [0070]).
Although Blanch discloses determining, one or more candidate models from a first type of models and a second type of models and selecting a chroma prediction model from the one or more candidate models, it is noted Blanch differs from the present invention in that it fails to particularly disclose based on a decision rule comprising comparing a coding parameter associated with a chroma block with a threshold. Li however, teaches based on a decision rule comprising comparing a coding parameter associated with a chroma block with a threshold (e.g. see linear model (LM) method assumption is inaccurate for complex content or large blocks and restricts the prediction accuracy, Abstract; Fig. 2 illustrates and suggests to select proposed mode, i.e. neural network mode, for regions with rich textures or structures and large blocks and select LM for smaller blocks (and less rich in textures or structures), see page 1800, paragraph before conclusion, which is according to rate-distortion cost criterion, see Section 3.2; thus, it would be obvious for a PHOSITA to take into account block sizes, number of pixels, depth and ratio (which corresponds to large blocks), as well as, quantization parameters (which corresponds to textures/complexity) into Blanch for determining between traditional angular modes, LM models or the disclosed neural network mode and to use thresholds to objectively judge which blocks are large and/or have complex contents; for example, by comparing block sizes with a threshold).
Therefore, given the teachings as a whole, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, having the references of Blanch and Li before him/her, to modify the chroma intra prediction in video coding and decoding of Blanch with the teachings of Li in order to increase prediction accuracy.
Further, although Blanch in view of Li teaches determining, based on a decision rule comprising comparing a coding parameter associated with a chroma block with a threshold, one or more candidate models from a first type of models and second type of models, it is noted Blanch differs from the present invention in that it fails to particularly disclose the decision rule indicated in a syntax structure as a syntax element. Deng however, teaches the decision rule indicated in a syntax structure (e.g. see syntax structure, e.g. see at least paragraph [0068]) as a syntax element (e.g. see identifying the model used is signaled by one or more syntax element(s), e.g. see at least paragraphs [485]; also see quadtree syntax of the CTU, e.g. see at least paragraph [0105]).
Therefore, given the teachings as a whole, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, having the references of Blanch, Li and Deng before him/her, to incorporate the teachings of Deng into the chroma intra prediction in video coding and decoding of Blanch as modified by Li in order to signal information in a structured way and increase prediction accuracy.
Regarding claim 2, Blanch further discloses wherein the reference signals of the chroma block comprise one or more of: a reconstruction of a luma block corresponding to the chroma block; an upper adjacent line of the luma block; a left adjacent line of the luma block; an upper adjacent line of the chroma block; or a left adjacent line of the chroma block (e.g. see at least one or more of reconstructed samples or reference array on the left of and above the current block, e.g. see at least paragraphs [0020]-[0021], [0076]).
Regarding claim 3, Blanch further discloses wherein the first type of models comprises one or more linear models that determine chroma prediction values based on linear coefficients and corresponding samples in the reconstruction of the luma block (e.g. see at least CCLM, e.g. see at least paragraph [0016]).
Regarding claim 5, Blanch in view of Li further teaches wherein the coding parameter comprises one or more of: a horizontal size of the chroma block; a vertical size of the chroma block; a number of pixels of the chroma block; a depth of the chroma block in a coding tree structure; a ratio of a width to a height of the chroma block; or a quantization parameter (Li: e.g. see linear model (LM) method assumption is inaccurate for complex content or large blocks and restricts the prediction accuracy, Abstract; Fig. 2 illustrates and suggests to select proposed mode, i.e. neural network mode, for regions with rich textures or structures and large blocks and select LM for smaller blocks (and less rich in textures or structures), see page 1800, paragraph before conclusion, which is according to rate-distortion cost criterion, see Section 3.2; thus, it would be obvious for a PHOSITA to take into account block sizes, number of pixels, depth and ratio (which corresponds to large blocks), as well as, quantization parameters (which corresponds to textures/complexity) into Blanch for determining between traditional angular modes, LM models or the disclosed neural network mode and to use thresholds to objectively judge which blocks are large and/or have complex contents; for example, by comparing block sizes with a threshold). The motivation above in the rejection of claim 1 applies here.
Regarding claim 6, Blanch in view of Li further teaches wherein the second type of models comprise one or more non-linear models (e.g. see traditional angular modes, LM models or the disclosed neural network mode, e.g. see at least paragraphs [0016], [0021], [0072]; neural network mode is non-linear type) and wherein the determining the one or more candidate models further comprises selecting a candidate from the second type of models based on: the horizontal size of the chroma block being greater than a first threshold and the vertical size of the chroma block being greater than a second threshold; the number of pixels of the chroma block being greater than a third threshold; a maximum of the ratio and a reciprocal of the ratio being less than a fourth threshold; or the quantization parameter being less than a fifth threshold (Li: e.g. see linear model (LM) method assumption is inaccurate for complex content or large blocks and restricts the prediction accuracy, Abstract; Fig. 2 illustrates and suggests to select proposed mode, i.e. neural network mode, for regions with rich textures or structures and large blocks and select LM for smaller blocks (and less rich in textures or structures), see page 1800, paragraph before conclusion, which is according to rate-distortion cost criterion, see Section 3.2; thus, it would be obvious for a PHOSITA to take into account block sizes, number of pixels, depth and ratio (which corresponds to large blocks), as well as, quantization parameters (which corresponds to textures/complexity) into Blanch for determining between traditional angular modes, LM models or the disclosed neural network mode and to use thresholds to objectively judge which blocks are large and/or have complex contents; for example, by comparing block sizes with a threshold). The motivation above in the rejection of claim 1 applies here.
Regarding claim 7, Blanch in view of Li further teaches wherein the second type of models comprise one or more linear models (e.g. see traditional angular modes, LM models or the disclosed neural network mode, e.g. see at least paragraphs [0016], [0021], [0072]; LM models are linear type) and wherein the determining the one or more candidate models further comprises selecting a candidate model from the first type of models based on: the horizontal size of the chroma block being less than a first threshold or the vertical size of the chroma block being less than a second threshold; the number of pixels of the chroma block being less than a third threshold; a maximum of the ratio and a reciprocal of the ratio being greater than a fourth threshold; or the quantization parameter being greater than a fifth threshold (Li: e.g. see linear model (LM) method assumption is inaccurate for complex content or large blocks and restricts the prediction accuracy, Abstract; Fig. 2 illustrates and suggests to select proposed mode, i.e. neural network mode, for regions with rich textures or structures and large blocks and select LM for smaller blocks (and less rich in textures or structures), see page 1800, paragraph before conclusion, which is according to rate-distortion cost criterion, see Section 3.2; thus, it would be obvious for a PHOSITA to take into account block sizes, number of pixels, depth and ratio (which corresponds to large blocks), as well as, quantization parameters (which corresponds to textures/complexity) into Blanch for determining between traditional angular modes, LM models or the disclosed neural network mode and to use thresholds to objectively judge which blocks are large and/or have complex contents; for example, by comparing block sizes with a threshold). The motivation above in the rejection of claim 1 applies here.
Regarding claims 10-12, 14-18 and 20, the claims recite analogous limitations to the claims above and are therefore rejected on the same premise.
Claim(s) 4, 13 and 19 is/are rejected under 35 U.S.C. 103 as being unpatentable over US 2023/0062509 A1 (“Blanch”) in view of A Hybrid Neural Network for Chroma Intra Prediction (“Li”) in further view of US 2022/0248025 A1 (“Deng”) in further view of US 2022/0337824 A1 (“Chen”).
Regarding claim 4, although Blanch discloses wherein the second type of models comprise one or more non-linear models that comprise: one or more layers configured to receive the reference signals of the chroma block and to generate a score distribution; and an layer configured to receive the score distribution and to generate the prediction of the chroma block (e.g. see Fig. 7 showing layers that receive reference samples and predict chroma based on generated probability), it is noted Blanch differs from the present invention in that it fails to particularly disclose hidden layers and an output layer. Chen however, teaches hidden layers and an output layer (e.g. see at least hidden layers and output layer in at least Figs. 9B-9C).
Therefore, given the teachings as a whole, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, having the references of Blanch, Li, Deng and Chen before him/her, to incorporate the teachings of Chen into the chroma intra prediction in video coding and decoding of Blanch as modified by Li and Deng in order to utilize popular neural network architecture for image/video applications.
Regarding claims 13, 19, the claims recite analogous limitations to the claims above and are therefore rejected on the same premise.
Claim(s) 9 is/are rejected under 35 U.S.C. 103 as being unpatentable over US 2023/0062509 A1 (“Blanch”) in view of A Hybrid Neural Network for Chroma Intra Prediction (“Li”) in further view of US 2022/0248025 A1 (“Deng”) in further view of US 2023/0047271 A1 (“Chubach”).
Regarding claim 9, although Blanch discloses wherein the second type of models comprise one or more non-linear models, it is noted Blanch differs from the present invention in that it fails to particularly disclose the method further comprising receiving, in a bitstream, a set of parameters for the one or more non-linear models. Chubach however, teaches the method further comprising receiving, in a bitstream, a set of parameters for the one or more non-linear models (e.g. see parameters signaled to the decoder in a bitstream, e.g. see at least paragraph [0034]).
Therefore, given the teachings as a whole, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, having the references of Blanch, Li, Deng and Chubach before him/her, to incorporate the teachings of Chubach into the chroma intra prediction in video coding and decoding of Blanch as modified by Li and Deng in order to dynamically update the neural network to improve coding efficiency.
Allowable Subject Matter
Claims 8 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.
Response to Arguments
Applicant’s arguments with respect to claim(s) 1-20 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.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
US 20230095387 A1, Dumas et al., Neural network-based intra prediction for video encoding or decoding
US 20220400272 A1, Lin et al., Content-adaptive online training for DNN-based cross component prediction with scaling factors
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 FRANCIS G GEROLEO whose telephone number is (571)270-7206. The examiner can normally be reached M-F 7:00 am - 3:30 pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Anna M Momper can be reached at (571) 270-5788. 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.
/Francis Geroleo/Primary Examiner, Art Unit 3619
Prosecution Insights