DETAILED ACTION
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 1/14/2026 has been entered.
Response to Arguments
Applicant has amended claims 21, 29, and 37; claims 21-40 are currently pending.
Applicant’s arguments filed 1/6/2026, with respect to the 35 U.S.C. 103 rejection(s) of claim(s) 21, 29, and 37 as being unpatentable over Roshtakhari (Pub. No. US20180336704A1) in view of Chen (Chen J, Little JJ. Sports camera calibration via synthetic data. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition workshops 2019 (pp. 0-0).) have been fully considered and are persuasive. Therefore, the rejection has been withdrawn. However, upon further consideration, a new ground(s) of rejection is made in view of previously cited reference, Micheloni (Micheloni C, Rinner B, Foresti GL. Video analysis in pan-tilt-zoom camera networks. IEEE Signal Processing Magazine. 2010 Sep 2;27(5):78-90.) as is detailed in the art rejection below. Therefore this action is made NON-FINAL.
Information Disclosure Statement
The information disclosure statement(s) (IDS) submitted on 1/21/2026 is/are in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement(s) has/have been considered by the examiner.
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) 21- 40 is/are rejected under 35 U.S.C. 103 as being unpatentable over Roshtakhari (Pub. No. US20180336704A1) in view of Chen (Chen J, Little JJ. Sports camera calibration via synthetic data. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition workshops 2019 (pp. 0-0).) and Micheloni (Micheloni C, Rinner B, Foresti GL. Video analysis in pan-tilt-zoom camera networks. IEEE Signal Processing Magazine. 2010 Sep 2;27(5):78-90.).
Regarding claim 29, Roshtakhari discloses A system for generating a fully trained calibration model, comprising: a processor; and a memory having programming instructions stored thereon, which, when executed by the processor, performs one or more operations, comprising: (Roshtakhari Claim 19; a device with a processor and memory with computer executable instructions is disclosed.) retrieving one or more training data sets from one or more data stores, wherein the one or more training data sets include a plurality of images captured by a camera system during a sporting event; (Roshtakhari ¶89; ground truth homographies for frames of broadcast videos of sporting events including basketball and hockey are disclosed.) generating a plurality of camera pose templates (Roshtakhari ¶80-81; a key-frame database is built, consisting of images with common characteristics like similar viewing angles and similar parts of the scene.) and a corresponding plurality of homography matrices (Roshtakhari ¶86; a database of transformation pairs can be generated using methods described. ¶65 a new frame is matched with a key-frame from the database and the corresponding homography of the key-frame is used as an initialization point.) refining at least one of the plurality of homography matrices based on a relative transformation between at least one of the plurality of camera pose templates and the corresponding at least one of the plurality of images; (Roshtakhari ¶70, ¶73 and equations (7) and (9); image to image homographies are estimated and the combination of equations (7) (the homography from previous frame or key-frame if currently processing a first frame in a time sequence see ¶65-66) and (9) (the homography between the previous frame/ key-frame and the currently processed frame). Therefore the difference in camera view between the key-frame (camera pose template) and the current frame is used to refine the current frame’s homography estimate.) and outputting a trained prediction model based on the trained neural network. (Roshtakhari ¶88-89; the trained network is utilized to calculate homographies of broadcast videos and the accuracy is evaluated therefore the trained network was output.)
Roshtakhari discloses a trained neural network (¶86-87), but not explicitly training a neural network to calibrate a camera based on the one or more training data sets the plurality of camera pose templates, and the plurality of refined homography matrices.
Chen, however, discloses training a neural network to calibrate a camera based on the one or more training data sets the plurality of camera pose templates, and the plurality of refined homography matrices (Chen Section I. Introduction ¶4-6 and Fig. 2; general “default” settings are used as a key-point starting place for training images. These “rough” estimates can then be refined and then used to train the model (see Fig. 2 Siamese network to feature extraction for the training path.) Section II. Related Work ¶2 explains that the camera calibration is equivalent to estimating the homography.)
It would have been obvious, before the effective filing date of the claimed invention, to one of ordinary skill in the art to modify the system of Roshtakhari with the teachings of Chen by including refined camera pose (i.e. refined homographies) in order to have realistic assumptions of camera locations in broadcast video camera calibration (Chen Section I. Introduction ¶5).
Although Roshtakhari discloses building the key-frame database using Girdhar and Dudek(Roshtakhari ¶83), the combination of Roshtakhari and Chen does not explicitly disclose generating a camera pose dictionary, wherein the generating includes determining a pan range, a tilt range, a focal length, or a camera location from the one or more training sets and sampling a range of possible camera poses; generating a plurality of camera pose templates and a corresponding plurality of homography matrices based on the camera pose dictionary
Micheloni, however, discloses generating a camera pose dictionary, wherein the generating includes determining a pan range, a tilt range, a focal length, or a camera location from the one or more training sets and sampling a range of possible camera poses (Micheloni slides 39, 46-47; the possible (range) pan and tilt positions are sampled and a look up table is created for the sampled positions.) generating a plurality of camera pose templates and a corresponding plurality of homography matrices based on the camera pose dictionary (Micheloni slides 39, 46-47; a look up table is created for possible pan and tilt angles. The rectification transformations/ homographies are computed and stored in the look up table.)
It would have been obvious, before the effective filing date of the claimed invention, to one of ordinary skill in the art to modify the system of the combination of Roshtakhari and Chen with the teachings of Micheloni by including uniform sampling of camera parameters in building the key-frame database in order to localize views in 3D while utilizing precalculated values of likely camera views reducing the run time processing power necessary.
Regarding claim 30, the combination of Roshtakhari, Chen, and Micheloni disclose the claim limitations with regards to claim 29, as described above. They further disclose wherein the plurality of camera pose templates are generated based on a high grid resolution approach. (Micheloni slides 46-47; creating a look up table (LUT) of camera views by sampling pan tilt ranges of PTZ camera and calculating corresponding transform pairs (homographies) is disclosed)
Regarding claim 31, the combination of Roshtakhari, Chen, and Micheloni disclose the claim limitations with respect to claim 30, as described above. They further disclose wherein the high grid resolution approach comprises setting a pan resolution, a tilt resolution, and a focal length resolution. (Micheloni slides 46-47; creating a look up table (LUT) of camera views by sampling pan tilt ranges of PTZ camera at equal intervals. Multiple focal lengths can be seen in slides15-18 and 51-52.)
Regarding claim 32, the combination of Roshtakhari, Chen, and Micheloni disclose the claim limitations with respect to claim 30, as described above. They further disclose wherein the trained neural network comprises: a semantic segmentation module; (Roshtakhari ¶68-69; identifying the planar surface of the field/ sporting event arena is disclosed. This includes detected lines and edges.) a camera pose initialization module; (Roshtakhari ¶65; the initial camera pose is estimated from the key-frame database.) and a homography refinement module. (Roshtakhari ¶70, ¶73 and equations (7) and (9); image to image homographies are estimated and the combination of equations (7) (the homography from previous frame or key-frame if currently processing a first frame in a time sequence see ¶65-66) and (9) (the homography between the previous frame/ key-frame and the currently processed frame). Therefore the difference in camera view between the key-frame (camera pose template) and the current frame is used to refine the current frame’s homography estimate.)
Regarding claim 33, the combination of Roshtakhari, Chen, and Micheloni disclose the claim limitations with respect to claim 32, as described above. They further disclose wherein the training the neural network is performed module-by-module. (Roshtakhari ¶68-69; the planar surface is identified using a learn color model (i.e. is trained) this is utilized in later steps. ¶85-86 key frame database is utilized in learning the refined homography.)
Regarding claim 34, the combination of Roshtakhari, Chen, and Micheloni disclose the claim limitations with respect to claim 32, as described above. They further disclose wherein the semantic segmentation module is configured to generate a semantic map. (Roshtakhari ¶68-69; identifying the planar surface of the field/ sporting event arena is disclosed. This includes detected lines and edges.)
Regarding claim 35, the combination of Roshtakhari, Chen, and Micheloni disclose the claim limitations with respect to claim 34, as described above. They further disclose wherein the camera pose initialization module is configured to determine a template of the plurality of camera pose templates for generating a homography matrix based on the semantic map. (Roshtakhari ¶81-82; a closest key-frame to the current frame is found by using similar features, including edges (semantic map). ¶86 the key-frames can be used in the calculation of the homography.)
Regarding claim 36, the combination of Roshtakhari, Chen, and Micheloni disclose the claim limitations with respect to claim 34, as described above. They further disclose wherein the homography refinement module is configured to generate a homography matrix based on a template of the plurality of camera pose templates and the semantic map. (Roshtakhari ¶73-74; feature based refinement (including edges i.e. semantic map) to the homography is performed. See also equation (10). ¶85-86 explain that the homography can be estimated using the key-frame database which would then be refined as in ¶73-74.)
Claims 21-28 are the corresponding method claims to claims 29-36 respectively and are rejected for similar reasons.
Claims 37-40 are the corresponding computer readable medium claims to claims 29-32 and are rejected for similar reasons.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to MEREDITH TAYLOR whose telephone number is (571)270-5805. The examiner can normally be reached M-Th 7:30-5.
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, Vincent Rudolph can be reached at (571)272-8243. 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.
/MEREDITH TAYLOR/Examiner, Art Unit 2671
/VINCENT RUDOLPH/Supervisory Patent Examiner, Art Unit 2671
Prosecution Insights