Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Arguments/Amendments
The amendment filed 06/18/2026 in response to the Non-Final Office Action mailed on 05/21/2026 has been entered.
Claims 1-20 are currently pending in U.S. Patent Application No. 18/304,651.
Applicant’s remarks filed 06/18/2026 have been fully considered and responded to below.
The rejections under 35 U.S.C. 112(b) are removed in view of the claim amendments.
Regarding the 35 U.S.C. 103 rejections, the Applicant’s remarks have been considered but are not persuasive. The Applicant asserts that “Maisa does not fit a ground truth ITM curve to paired SDR and HDR training samples, and does not train its model based on parameters of any such curve.”. The Examiner respectfully disagrees. The presented claim amendments disclose a ML model that is trained based on “one or more parameters of a ground truth ITM curve”, where the ITM curve is fit to a pair of SDR and corresponding HDR images. Maisa trains an ML model (the multilinear regression model) through an iterative stepwise regression, which iteratively modifies the multilinear regression model with different independent variables (i.e., statistics information from an image) by computing F-statistics, which involves a comparison between the predicted γ values obtained from the regression model with ground-truth γ values (i.e., parameters of a ground truth ITM curve). Therefore, the machine learning model disclosed by Maisa is trained based on one or more parameters (i.e., the γ values) of an ITM curve. Furthermore, the ground-truth γ values are obtained by asking individuals to manually tweak a γ curve (i.e., a ground truth ITM curve) for a SDR image to obtain a “best rendition” HDR image (i.e., fitting an ITM curve between a pair of SDR and corresponding HDR images). The Applicant further notes that the additional references presented in the non-final rejection do not disclose the amended limitations, however this point is moot as the amended limitation is disclosed by Maisa.
Additionally, due to the change of scope necessitated by the Applicant’s amendments, a new ground of rejection will also be presented below.
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.
Claims 1, 11, and 17 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Luzardo et al. (“Fully-automatic inverse tone mapping algorithm based on dynamic mid-level tone mapping”, DOI: 10.1017/ATSIP.2020.5, Publication Year: 2020; hereinafter “Luzardo”).
Regarding Claim 1, Luzardo discloses a method comprising (see III. Proposed Expansion Function):
receiving, as input, standard dynamic range (SDR) content (V. Results, Luzardo discloses obtaining low dynamic range (LDR) images from a final test dataset.);
obtaining statistics information corresponding to the SDR content (A. HDR-LDR Image Dataset, Table 2, Luzardo discloses computing first-order image statistics for an LDR image.);
inputting the statistics information into a machine learning model configured to output one or more parameters of an inverse tone-mapping (ITM) curve (V. Results, IV. Fully-Automatic Inverse Tone Mapping, C) Multi-linear regression, Equation 6, Fig. 1, Luzardo discloses a multi-linear regression model (i.e., a machine learning model) which takes input of various image statistics to output a middle-gray value
m
o
, which is a parameter of the proposed inverse tone mapping function.) , wherein the machine learning model is trained based on one or more parameters of a ground truth ITM curve, the ground truth ITM curve fit to one or more SDR training samples and one or more corresponding high dynamic range (HDR) training samples (IV. Fully-Automatic Inverse Tone Mapping, C) Multi-linear regression, Luzardo discloses training a multi-linear regression model using a step-wise regression method for variable selection. The Examiner notes that the stepwise regression method involves iteratively testing different variable combinations to identify the ideal multi-linear regression model. This iterative process involves calculating F-values, which is based on a comparison between the predicted
m
o
value and the ground-truth
m
o
value (i.e., a parameter of a ground truth ITM curve). The ground-truth
m
o
values are obtained from a subjective study (see B. Subjective study), wherein users are asked to adjust a
m
o
value starting from an SDR image until it matched a corresponding HDR image.);
determining the one or more parameters of the ITM curve based on the output of the machine learning model; and converting the SDR content to high dynamic range (HDR) content using the ITM curve, wherein the resulting HDR content is provided to a display device for presentation (IV. Fully-Automatic Inverse Tone Mapping, V. Results, Luzardo discloses inputting the image statics from an LDR image into a multi-linear regression model to obtain a predicted
m
o
value, which is used to estimate inverse tone mapping parameters to generate an HDR image. The Examiner notes the output HDR images presented in Fig. 4 as well as defining display resolution parameters.).
Claims 11 and 17 are the system and non-transitory processor-readable medium claims corresponding to claim 1 and are similarly rejected (the Examiner notes that the disclosure presented by Luzardo involves complex image processing and consequent displaying of image content, which requires a computing device consisting of the claimed system and non-transitory process-readable medium components.).
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.
Claims 1-2, 4, 10-11, and 16-17 are rejected as being unpatentable over Maisa et al. (“Dynamic range expansion based on image statistic”, DOI: https://doi.org/10.1007/s11042-015-3036-0; hereinafter “Maisa”) in view of Su et al. (US 2022/0301124; hereinafter “Su”).
Regarding Claim 11, Maisa discloses A method comprising (see 3. Multilinear models for gamma expansion.):
receiving, as input, standard dynamic range (SDR) content (3.1 Input Data, Maisa teaches obtaining images from an SDR image database.);
obtaining statistics information corresponding to the SDR content (3.1. Image Data, Maisa teaches determining a variety of image statistics based on linearized luminance values.);
inputting the statistics information into a machine learning model configured to output one or more parameters of an inverse tone-mapping (ITM) curve, wherein the machine learning model is trained based on one or more parameters of a ground truth ITM curve, the ground truth ITM curve fit to one or more SDR training samples and one or more corresponding high dynamic range (HDR) training samples (3.1. Image Data, Equation 4, Maisa discloses obtaining a multilinear regression model (i.e., a machine learning model), which takes input different image statistics information (Lh (geometric mean luminance), k1 (image key based on maximum and minimum luminance values), and pov (percentage of overexposed pixels).). The multilinear regression model outputs a predicted γ value used to perform gamma expansion used for reverse tone-mapping. The Examiner notes the Examiner’s remarks made above regarding how Maisa’s use of a multilinear regression model disclosed the claimed limitations.).
determining, the one or more parameters of the ITM curve based on the output of the machine learning model (3.2. Multilinear regression, Maisa teaches utilizing different image statistics to generate a multilinear model (represented by Equation 4) which predicts a γ value used to perform gamma expansion used for reverse tone-mapping.); and
converting the SDR content to high dynamic range (HDR) content using the ITM curve (4. Results, Figs. 5, 8, 9, Maisa teaches utilizing the multilinear model to perform gamma expansion to convert a SDR image to HDR.),
Maisa does not explicitly disclose and converting the SDR content to high dynamic range (HDR) content using the ITM curve (italicized for context), wherein the resulting HDR content is provided to a display device for presentation (The Examiner notes 1. Introduction, 4. Results, Maisa teaches a SDR to HDR image conversion process using a reverse tone mapping operator. The Examiner asserts that the HDR images produced by Maisa (shown in Figs. 8-9), are performed on a computer, and as they are then furthermore visually analyzed, Maisa’s methods involve the claimed “resulting content is provided to a display device”. This is supported by the disclosure in 1. Introduction, wherein Maisa discloses that reverse tone mapping operation taught by Maisa “is simple enough that modern displays can compute it on the fly”. However, for the record the Examiner notes below another reference teaching the claimed limitation.).
Su teaches and converting the SDR content to high dynamic range (HDR) content using the ITM curve (italicized for context), wherein the resulting HDR content is provided to a display device for presentation ([0192], Fig.4B, Su teaches converting an SDR image into an HDR image and presenting the image on a display device.).
Maisa and Su are considered to be analogous to the claimed invention as they are in the same field of SDR to HDR image conversion. Therefore, it would have been obvious to one having ordinary skill in the art before the effective filling date of the claimed invention to modify the invention of Maisa by incorporating the HDR display capabilities as taught by Su. The motivation for this combination being the ability to display an image to a user at the end of processing.
Claims 11 and 17 are the system and non-transitory processor-readable medium claims corresponding to claim 1 and are similarly rejected (see [0193-1934], Su).
Regarding Claim 2, Maisa in view of Su teaches the method of claim 1, wherein the display device has HDR rendering capabilities ([0020], [0192], Fig. 1, Fig. 4B, Su teaches an HDR display with the ability to display HDR images.).
Regarding Claim 4, Maisa in view of Su teaches the method of claim 1.
The current combination of Maisa in view of Su does not explicitly teach wherein obtaining statistics information corresponding to the SDR comprises: parsing metadata corresponding to the SDR content from SDR signals of the SDR content, wherein the metadata comprises the statistics information.
Su further teaches obtaining statistics information corresponding to the SDR comprises: parsing metadata corresponding to the SDR content from SDR signals of the SDR content, wherein the metadata comprises the statistics information ([0023-0024], Su teaches utilizing SDR image metadata to generate an HDR image. The Examiner notes that Su’s metadata includes composter data to generate backward reshaping mappings, which is analogous to the claimed “statistics information” as both the claimed “statistics information” and composer data are used in the process to generate backward reshaping mappings (i.e., an ITM curve).).
It would have been obvious to one having ordinary skill in the art before the effective filling date of the claimed invention to further modify the invention of Maisa in view of Su such that it includes the incorporation and processing of metadata as taught by Su. The motivation for this combination being the ability to incorporate additional information in the formation of an HDR image from an SDR image.
Regarding Claim 10, Maisa in view of Su teaches the method of claim 1, wherein the machine learning model is implemented in a Digital Signal Processor (DSP) or a central processing unit (CPU) of the display device ([0017-0018], [0036], [0193], Su teaches a display device comprising a processor where machine learning models are generated and stored.).
Claim 16 is the system claim corresponding to claim 10 and are similarly rejected (see [0193-1934], Su).
Claims 3, 12, and 18 are rejected as being unpatentable over Maisa in view of Su in view of Kuo et al. (“Content-adaptive inverse tone mapping”, DOI: 10.1109/VCIP.2012.6410798; hereinafter “Kuo”).
Regarding Claim 3, Maisa in view of Su teaches the method of claim 1.
Maisa in view of Su does not explicitly teach wherein the statistics information comprises, for each SDR image of the SDR content, at least one of a histogram of the SDR image or linear luminance percentiles sampled from a cumulated distribution function (CDF) of the SDR image based on pre-defined sampling percentage values.
Kuo teaches wherein the statistics information comprises, for each SDR image of the SDR content, at least one of a histogram of the SDR image (Section 3.1.2. Scene Classifier, Kuo teaches extracting image features from training data to input into a support vector machine (SVM), of which one of the features is a histogram.) or linear luminance percentiles sampled from a cumulated distribution function (CDF) of the SDR image based on pre-defined sampling percentage values.
Maisa, Su, and Kuo are considered to be analogous to the claimed invention as they are in the same field of SDR to HDR image conversion. Therefore, it would have been obvious to one having ordinary skill in the art before the effective filling date of the claimed invention to modify the invention of Maisa in view of Su such that the percentile statistics calculated by Maisa in view of Su are obtained based on the histogram taught by Kuo. The motivation for this combination being the ability to generate a distribution of luminance values, which can then be used to sample specific percentiles which can be used for further downstream calculations.
Claims 12 and 18 are the system and non-transitory processor-readable medium claims corresponding to claim 3 and are similarly rejected (see [0193-1934], Su).
Claim 5 is rejected as being unpatentable over Maisa in view of Su in view of Wen et al. (US2018/0098094; hereinafter “Wen”) in view of Kim et al. (US2008/0002062; hereinafter “Kim”.).
Regarding Claim 5, Maisa in view of Su teaches the method of claim 1.
Maisa in view of Su does not teach wherein obtaining statistics information corresponding to the SDR content comprises: for each SDR image of the SDR content: calculating a histogram of the SDR image; calculating a cumulated distribution function (CDF) of the SDR image based on the histogram of the SDR image; and sampling linear luminance percentiles from the CDF of the SDR image based on pre-defined sampling percentage values.
Wen teaches obtaining statistics information corresponding to the SDR content comprises: for each SDR image of the SDR content: calculating a histogram of the SDR image ([0066-0067], Wen teaches calculating an SDR histogram.); calculating a cumulated distribution function (CDF) of the SDR image based on the histogram of the SDR image ([0146], Fig. 4A, Wen teaches constructing a CDF of an SDR image based on an SDR histogram.);
Maisa, Su, and Wen are considered to be analogous to the claimed invention as they are in the same field of SDR to HDR image conversion. Therefore, it would have been obvious to one having ordinary skill in the art before the effective filling date of the claimed invention to modify the invention of Maisa in view of Su such that the histogram obtained by Maisa in view of Su is used to calculate a CDF of the SDR image as taught by Wen. The motivation for this combination being the ability to use the CDF for further downstream processing and calculations.
Maisa in view of Su in view of Wen does not teach sampling linear luminance percentiles from the CDF of the SDR image based on pre-defined sampling percentage values.
Kim teaches sampling linear luminance percentiles from the CDF of the SDR image based on pre-defined sampling percentage values ([0033], Fig. 2, Kim teaches sampling a high range and low range threshold point from a luminance CDF graph, wherein the threshold points can be defined at 10% and 90%.).
Maisa, Su, Wen, and Kim are considered to be analogous to the claimed invention as they are in the same field improving image quality. Therefore, it would have been obvious to one having ordinary skill in the art before the effective filling date of the claimed invention to modify the invention of Maisa in view of Su in view of Wen such that the CDF was sampled in a similar process as described by Kim. The motivation for this combination being the ability to utilize the CDF and obtain pertinent values relating to the SDR image which are beneficial for further downstream processing.
Claim 6 is rejected as being unpatentable over Maisa in view of Su in view of Wen.
Regarding Claim 6, Maisa in view of Su teaches the method of claim 1.
Maisa in view of Su does not teach wherein the ITM curve is an n-th order polynomial curve.
Wen teaches wherein the ITM curve is an n-th order polynomial curve ([0081], Fig. 4B, Wen teaches using a set of polynomial pieces to approximate a backward reshaping function to convert SDR images to HDR images.).
It would have been obvious to one having ordinary skill in the art before the effective filling date of the claimed invention to modify the invention of Maisa in view of Su such the inverse tone mapping operator taught by Maisa in view of Su is the set of polynomial pieces used to approximate a backward reshaping function as taught by Wen. The motivation for this combination being the ability to define an explicit function and its individual components which can be used for SDR to HDR conversion.
Claims 7, 13, and 19 are rejected as being unpatentable over Maisa in view of Su in view of Wen in view of Yeung et al. (US2020/0043125; hereinafter “Yeung”).
Regarding Claim 7, Maisa in view of Su in view of Wen teaches the method of claim 6.
Maisa in view of Su in view of Wen does not teach wherein the n-th order polynomial curve is one of a Bernstein polynomial curve or a Bézier curve.
Yeung teaches wherein the n-th order polynomial curve is one of a Bernstein polynomial curve or a Bézier curve ([0046-0047], Yeung teaches using a Bézier curve for tone-mapping.).
Maisa, Su, Wen, and Yeung are considered to be analogous to the claimed invention as they are in the same field improving image quality. It would have been obvious to one having ordinary skill in the art before the effective filling date of the claimed invention to modify the invention of Maisa in view of Su in view of Wen such that the polynomial curve taught by Maisa in view of Su in view of Wen is the Bézier curve taught by Yeung. The motivation for this combination being the ability to use a smooth defined curve which can be mathematically defined.
Regarding Claim 13, Maisa in view of Su teaches the system of claim 11.
Maisa in view of Su does not teach wherein the ITM curve is an n-th order polynomial curve, and the n-th order polynomial curve is one of a Bernstein polynomial curve or a Bézier curve.
Wen teaches wherein the ITM curve is an n-th order polynomial curve ([0081], Fig. 4B, Wen teaches using a set of polynomial pieces to approximate a backward reshaping function to convert SDR images to HDR images.),
It would have been obvious to one having ordinary skill in the art before the effective filling date of the claimed invention to modify the invention of Maisa in view of Su such the inverse tone mapping operator taught by Maisa in view of Su is the set of polynomial pieces used to approximate a backward reshaping function as taught by Wen. The motivation for this combination being the ability to define an explicit function and its individual components which can be used for SDR to HDR conversion.
Maisa in view of Su in view of Wen does not teach and the n-th order polynomial curve is one of a Bernstein polynomial curve or a Bézier curve.
Yeung teaches and the n-th order polynomial curve is one of a Bernstein polynomial curve or a Bézier curve ([0046-0047], Yeung teaches using a Bézier curve for tone-mapping.).
Maisa, Su, Wen, and Yeung are considered to be analogous to the claimed invention as they are in the same field improving image quality. It would have been obvious to one having ordinary skill in the art before the effective filling date of the claimed invention to modify the invention of Maisa in view of Su in view of Wen such that the polynomial curve taught by Maisa in view of Su in view of Wen is the Bézier curve taught by Yeung. The motivation for this combination being the ability to use a smooth defined curve which can be mathematically defined.
Claim 19 is the non-transitory processor-readable medium claim corresponding to claim 13, and is similarly rejected (see [0193-1934], Su).
Claims 8 and 14 are rejected as being unpatentable over Maisa in view of Su in view of Kadu et al. (US2021/0350512; hereinafter “Kadu”).
Regarding Claim 8, Maisa in view of Su teaches the method of claim 1.
Maisa in view of Su does not teach wherein the machine learning model is trained offline.
Kadu teaches wherein the machine learning model is trained offline ([0059], Kadu teaches training a machine learning model offline.).
Maisa, Su, and Kadu are considered to be analogous to the claimed invention as they are in the same field processing SDR content. It would have been obvious to one having ordinary skill in the art before the effective filling date of the claimed invention to modify the invention of Maisa in view of Su in view of Wen such that the machine learning model used to produce parameters of an ITM curve was trained offline as taught by Kadu. The motivation for this combination being the ability to reduce the workload on the system such that portions of the computation can be performed offline instead of in real-time.
Claim 14 is the system corresponding to claim 8 and is similarly rejected (see [0193-1934], Su).
Conclusion
No claims are allowed. Claims 9, 15, and 20 are free of the prior art. Final allowability is contingent on overcoming all presented rejections.
THIS ACTION IS MADE FINAL. 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 PROMOTTO TAJRIAN ISLAM whose telephone number is (703)756-5584. The examiner can normally be reached Monday - Friday 8:30 am - 5:00 pm EST.
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, Chan Park can be reached at (571) 272-7409. 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.
/PROMOTTO TAJRIAN ISLAM/Examiner, Art Unit 2669 /CHAN S PARK/Supervisory Patent Examiner, Art Unit 2669
1 The Examiner notes that this 35 U.S.C. 103 rejection is presented in addition to the 35 U.S.C. 102(a)(1) rejection made above and will be the basis for the remaining rejections regarding the dependent claims made below.
Prosecution Insights