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 .
Allowable Subject Matter
Claims 1-19 are allowed.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claim 20 is rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claim 20 recites a video bitstream that operates as software. It is awkwardly stated and does not make sense.
The specification defines “bitstream” as video data encoded according to the disclosed encoding method, which can be encrypted and stored in memory. Spec. at [0317]. The bitstream is not software: it is data. Therefore, while it can be “decoded…,” i.e., decrypted under BRI, it cannot “enable[] a coding device to… generate a video comprising coded image data and a plurality of syntax elements” or the remaining functional steps of Claim 20—because it is not software. It is data.
Claim Rejections - 35 USC § 102
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claim(s) 20 is rejected under 35 U.S.C. 102(a)(2) as being anticipated by Aucsmith (US 5754658 A).
Regarding Claim 20, Aucsmith (US 5754658 A) discloses a non-transitory computer-readable storage medium storing an encoded bitstream (the encrypted video picture may be further encoded to generate an encrypted, compressed bitstream that may be stored, Column 9 lines 25-35). The remainder of Claim 20 has no patentable weight.
Reasons for Allowance
The following is an examiner’s statement of reasons for allowance:
The invention is a conditional variational auto-encoder (Spec. at [0159]), which encodes video into a bitstream (Spec. at [0004]), taking a quality parameter β as input (Spec. at [0211]). The quality parameter controls the rate-distortion tradeoff, i.e., the decoded image quality and the bitrate (Spec. at [0194]). The auto-encoder is designed with gain units (Spec. at [0194]) that convert the input quality parameter β into a gain vector, or weights, (Spec. at [0204]) that scale the video features to various bitrates/qualities, from low to high (Spec. at [0200]). The auto-encoder is trained on a set of gain vectors (Spec. at [0205]), and any β within the range that the auto-encoder was trained on can be converted into a gain vector by interpolation (Spec. at [0200]). This is known in the art.
The inventive concept is that the auto-encoder can take values of β outside the range that the auto-encoder was trained on and extrapolate additional gain vectors (Spec. at [0195]). The applicable claim language is “a value of the first coding parameter is smaller than a preset minimal value or larger than a preset maximal value,” (Claim 1). Examiner’s search has not yielded a reference that can be combined with existing published documents to teach or suggest this feature.
Conditional variational auto-encoders that can encode video into continuously adaptive bitrate are known in the art:
Choi, Y., El-Khamy, M., & Lee, J. (2019). Variable rate deep image compression with a conditional autoencoder. In Proceedings of the IEEE/CVF international conference on computer vision (pp. 3146-3154).
Guo, T., Wang, J., Cui, Z., Feng, Y., Ge, Y., & Bai, B. (2020). Variable rate image compression with content adaptive optimization. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (pp. 122-123).
Cui, Z., Wang, J., Gao, S., Guo, T., Feng, Y., & Bai, B. (2021). Asymmetric gained deep image compression with continuous rate adaptation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 10532-10541).
Zhang, S., Wang, L., Mao, X., Yang, F., & Wan, S. (2022, December). Rate controllable learned image compression based on rfl model. In 2022 IEEE International Conference on Visual Communications and Image Processing (VCIP) (pp. 1-5). IEEE.
These references are limited to quality parameters within the range of gain vectors that the auto-encoder was trained on. Patent documents on interpolating between the training set of gain vectors post-date the effective filing date of the application:
US 20240212221 A1
US 20250168412 A1
US 20250247552 A1
US 20250247542 A1
US 20250343935 A1
US 20250343917 A1
US 20250379990 A1
US 20250373827 A1
US 20260012642 A1
US 20260019577 A1
None of the above reference extrapolate beyond the training set of gain vectors.
There is a field of study in neural networks called out-of-distribution generalization. It refers to neural networks executing on data that the neural network was not trained on. There are several references in this area, but none of them consider compressing video to qualities outside the range of qualities that the neural network was trained on. References in this field include
Lei, E., Hassani, H., & Bidokhti, S. S. (2021). Out-of-distribution robustness in deep learning compression. arXiv preprint arXiv:2110.07007.
Na, G. S., & Park, C. (2022, August). Nonlinearity encoding for extrapolation of neural networks. In Proceedings of the 28th ACM SIGKDD conference on knowledge discovery and data mining (pp. 1284-1294).
Webb, T., Dulberg, Z., Frankland, S., Petrov, A., O’Reilly, R., & Cohen, J. (2020, November). Learning representations that support extrapolation. In International conference on machine learning (pp. 10136-10146). PMLR.
Bai, H., Zhou, F., Hong, L., Ye, N., Chan, S. H. G., & Li, Z. (2021). Nas-ood: Neural architecture search for out-of-distribution generalization. In Proceedings of the IEEE/CVF international conference on computer vision (pp. 8320-8329).
Examiner’s search has not yielded a supporting reference that can be combined with continuously adaptive variational auto-encoders known in the art to demonstrate that persons of ordinary skill in the would have found the claimed invention obvious.
Any comments considered necessary by applicant must be submitted no later than the payment of the issue fee and, to avoid processing delays, should preferably accompany the issue fee. Such submissions should be clearly labeled “Comments on Statement of Reasons for Allowance.”
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Johannes Balle et al., "VARIATIONAL IMAGE COMPRESSION WITH A SCALE HYPERPRIOR." arXiv:1802.01436v2 [eess.IV] 1 May 2018, total 23 pages.
Johannes Balle et al., "DENSITY MODELING OF IMAGES USING A GENERALIZED NORMALIZATION TRANSFORMATION," arXiv: 1511.06281v4 [cs.LG] 29 Feb 2016, 14 pages.
US 20150237375 A1 – dividing a block into one of four quantization grouping based on complexity/noise
US 20170019673 A1 – omitting decoding of a non-reference non-output layer in a scalabe coder
Any inquiry concerning this communication or earlier communications from the examiner should be directed to SHADAN E HAGHANI whose telephone number is (571)270-5631. The examiner can normally be reached M-F 9AM - 5PM.
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, Jay Patel can be reached at 571-272-2988. 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.
/SHADAN E HAGHANI/ Examiner, Art Unit 2485