DETAILED ACTION
Claims 1–20 are still pending in the instant application.
This Office Action is in response to the Applicant’s arguments filed on 09/09/2026.
THIS OFFICE ACTION IS MADE FINAL.
Response to Arguments
Examiner thanks Applicant filed an amendment to expedite patent prosecution to overcome the § 103 rejection. However, Examiner respectfully dissents that the instant application is in condition for allowance.
Examiner objected to claim 3 in the Office Action filed on 06/09/2026. In response to the Action, Applicant amended claim 1 to broadly recite claim 3. See his remarks p. 11.
First, Examiner decided that the amended claim language does not fully encompass original claim 3 where Applicant aggregates epipolar line corresponding to the target pose information in two input images. See his Spec. ¶55. The original claim explicitly demonstrates deterministic geometric operations, whereas the amendment replaces these concrete geometric steps with abstract functional results driven by a neural network. That is, the amendment introduces “epipolar geometric warping” and a conventional generation step of another image based on the warped features; to the contrary, sampling along an explicit line and aggregating related features is structurally distinct (emphasis added) from extracting, warping, and training thereof via a neural network model.
Second, Applicant fails to specifically disclose whether a second set of camera parameters are correlated to warped feature(s) in light of specification. Examiner referred to Applicant’s Spec. ¶53. The amended claim language implies that the first set of camera parameters are extracted by a neural network and the warped features are applied corresponding to a second set of camera parameters. But ¶53 suggests that Applicant’s warp operator 156 discloses geometric warping “based on the condition view image 1461 associated with the source image to determine a second set of camera parameters.” In the specification’s architecture, the second set of camera parameters is derived by using the condition view image as a geometric reference (i.e., finding the correspondence between the source view and the condition view).
Under MPEP § 2164(a), the specification must teach those skilled in the art how to use the full scope of the claimed invention without undue experimentation.2 ¶53 only teaches how epipolar line sampling is conducted using a condition view image and a differentiable aggregator; it does not teach how a neural network can map features to a second set of camera parameters via “epipolar geometric warping” without a foundational reference view.
In conclusion, a new ground of rejection in view of Hoon (KR 102110690 B1) has been arisen as Applicant amended the claimed language.
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, 5–6, 11–12, 16–17 and 19 are rejected under 35 U.S.C. § 103 as being unpatentable over Grossmann et al. (U.S. 9,338,439 B2) in view of Hoon (KR 102110690 B1).
Regarding claim 1, Grossman discloses one or more processor comprising:
one or more circuits to: generate, using at least one neural network, one or more features of a source image (one or more features of a source image construed as raw images) that depicts a scene using a first set of camera parameters (a first set of camera parameters construed as warped parameters). (Per Fig. 4 at step 410, Grossman discloses a pair of raw images. Then, he applies original warping parameters3 to determine whether the extent of epipolar misalignment is eminent. Grossman col. 7 line 51 – col. 8 line 4. [a]t 460, no adjustment is needed on the basis of the above epipolar misalignment determination. Note that while no adjustment may be needed on these grounds, additional testing (discussed below) may determine that adjustment of warping parameters is needed for other reasons.)
However, Grossman fails to specifically disclose apply an epipolar warping to the one or more features to determine one or more warped features at a second set of camera parameters, and generate, using the at least one neural network, a synthetic image representing the one or more warped features and corresponding to the second set of camera parameters.
In related art, Hoon discloses apply an epipolar geometric warping (epipolar geometric warping construed as warping image plane) to the one or more features to determine one or more warped features at a second of camera parameters (A second of camera parameters construed as characteristic parameters. See his Spec. ¶25. [t]he characteristic parameters are stored as intrinsic parameters (K) and extrinsic parameters (R, T), where the intrinsic parameter (K) represents the projection characteristics of the camera itself, and the extrinsic parameters (R, T) represent the rotation and translation position characteristics of the camera.); and (Hoon’s feature extraction unit 240 discloses a warping image plane in the warp feature map to capture multi-viewpoint images applying camera parameters. Ibid. ¶39. H x W is the size of the individual feature map, D is the size of the descriptor, and the eature of a pixel (p) in the warp feature map extracted from the nth warping image plane having D dimension is represented as…)
generate, using the at least one neural network, a synthetic image (a synthetic image construed as a reference viewpoint image) representing the one or more warped features and corresponding to the second set of camera parameters. (Hoon’s spatial transition unit 220 applies characteristic parameters to sample other viewpoint image such that other a plurality of warped images is generated. Ibid. ¶86. [u]ses the characteristic parameters (R, T, K) to compensate and sample the selected other viewpoint image so that it corresponds to the camera characteristic parameters of the reference viewpoint image to generate a plurality of warped images.)
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date to incorporate the teachings of Hoon into the teachings of Grossman to effectively reduce occlusion as a number of viewpoint increases. Ibid. ¶7.
Regarding claim 12, Grossman discloses a system, comprising:
one or more processing units; and (Fig. 16, a CPU 1620)
one or more memory units (Fig. 16, a memory 1610) storing instructions that, when executed by the one or more processing units, cause the one or more processing units to execute operations comprising:
generate, using at least one neural network, one or more features of a source image (one or more features of a source image construed as raw images) that depicts a scene using a first set of camera parameters (a first set of camera parameters construed as warped parameters). (Per Fig. 4 at step 410, Grossman discloses a pair of raw images. Then, he applies original warping parameters to determine whether the extent of epipolar misalignment is eminent. Grossman col. 7 line 51 – col. 8 line 4. [a]t 460, no adjustment is needed on the basis of the above epipolar misalignment determination. Note that while no adjustment may be needed on these grounds, additional testing (discussed below) may determine that adjustment of warping parameters is needed for other reasons.)
However, Grossman fails to specifically disclose apply an epipolar warping to the one or more features to determine one or more warped features at a second set of camera parameters, and generate, using the at least one neural network, a synthetic image representing the one or more warped features and corresponding to the second set of camera parameters.
In related art, Hoon discloses apply an epipolar geometric warping (epipolar geometric warping construed as warping image plane) to the one or more features to determine one or more warped features at a second of camera parameters (A second of camera parameters construed as characteristic parameters. See his Spec. ¶25. [t]he characteristic parameters are stored as intrinsic parameters (K) and extrinsic parameters (R, T), where the intrinsic parameter (K) represents the projection characteristics of the camera itself, and the extrinsic parameters (R, T) represent the rotation and translation position characteristics of the camera.); and (Hoon’s feature extraction unit 240 discloses a warping image plane in the warp feature map to capture multi-viewpoint images applying camera parameters. Ibid. ¶39. H x W is the size of the individual feature map, D is the size of the descriptor, and the eature of a pixel (p) in the warp feature map extracted from the nth warping image plane having D dimension is represented as…)
generate, using the at least one neural network, a synthetic image (a synthetic image construed as a reference viewpoint image) representing the one or more warped features and corresponding to the second set of camera parameters. (Hoon’s spatial transition unit 220 applies characteristic parameters to sample other viewpoint image such that other a plurality of warped images is generated. Ibid. ¶86. [u]ses the characteristic parameters (R, T, K) to compensate and sample the selected other viewpoint image so that it corresponds to the camera characteristic parameters of the reference viewpoint image to generate a plurality of warped images.)
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date to incorporate the teachings of Hoon into the teachings of Grossman to effectively reduce occlusion as a number of viewpoint increases. Ibid. ¶7.
Regarding claim 19, Grossman discloses a method, comprising:
generating, by one or more processors, using at least one neural network, one or more features of a source image (one or more features of a source image construed as raw images) that depicts a scene using a first set of camera parameters (a first set of camera parameters construed as warped parameters). (Per Fig. 4 at step 410, Grossman discloses a pair of raw images. Then, he applies original warping parameters to determine whether the extent of epipolar misalignment is eminent. Grossman col. 7 line 51 – col. 8 line 4. [a]t 460, no adjustment is needed on the basis of the above epipolar misalignment determination. Note that while no adjustment may be needed on these grounds, additional testing (discussed below) may determine that adjustment of warping parameters is needed for other reasons.)
However, Grossman fails to specifically disclose applying, by the one or more processors, a warping operation to the one or more features to determine one or more warped features at a second set of camera parameters; and generate by the one or more processors, using the at least one neural network, a synthetic image representing the one or more warped features and corresponding to the second set of camera parameters.
In related art, Hoon discloses applying, by the one or more processors, a warping operation to the one or more features to determine one or more warped features at a second set of camera parameters (A second of camera parameters construed as characteristic parameters. See his Spec. ¶25. [t]he characteristic parameters are stored as intrinsic parameters (K) and extrinsic parameters (R, T), where the intrinsic parameter (K) represents the projection characteristics of the camera itself, and the extrinsic parameters (R, T) represent the rotation and translation position characteristics of the camera.); and (Hoon’s feature extraction unit 240 discloses a warping image plane in the warp feature map to capture multi-viewpoint images applying camera parameters. Ibid. ¶39. H x W is the size of the individual feature map, D is the size of the descriptor, and the eature of a pixel (p) in the warp feature map extracted from the nth warping image plane having D dimension is represented as…)
generating, by the one or more processors, using the at least one neural network, a synthetic image (a synthetic image construed as a reference viewpoint image) representing the one or more warped features and corresponding to the second set of camera parameters. (Hoon’s spatial transition unit 220 applies characteristic parameters to sample other viewpoint image such that other a plurality of warped images is generated. Ibid. ¶86. [u]ses the characteristic parameters (R, T, K) to compensate and sample the selected other viewpoint image so that it corresponds to the camera characteristic parameters of the reference viewpoint image to generate a plurality of warped images.)
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date to incorporate the teachings of Hoon into the teachings of Grossman to effectively reduce occlusion as a number of viewpoint increases. Ibid. ¶7.
Regarding claim 5, Grossman as modified by Hoon, discloses the one or more processors, wherein the at least one neural network comprises a stable diffusion model. (Hoon discloses a Siamese Neural Network. Hoon Spec. ¶36. [e]ach of the multiple neural networks can be implemented as, for example, a Convolutional Neural Network (CNN), or as a Siamese Neural Network with the same structure and weights.)
Regarding claim 6, Grossman as modified by Hoon, discloses the one or more processors, wherein representations of the one or more features in at least one layer of at least one the neural network are unmodified by the epipoloar geometric warping. (Per Fig. 8, Grossman discloses adjusted warping parameters 820 after processing rectified images 350a–350c. Grossman col. 10 line 61 – col. 11 line 16. This module may receive rectified images 350a . . . c and produce adjusted warping parameters 820.)
Regarding claim 11, Grossman as modified by Hoon, discloses the one or more processors, wherein the one or more processors comprise at least one of: a system for generating synthetic data. (Per Fig. 1, Grossman discloses optical systems 110. Grossman col. 3 lines 19–35. [o]ne or more optical systems 110, such as lenses, imagers 120 that may digitize images formed by the lenses,)
Regarding claim 16, it has been rejected in the same manner as claim 5.
Regarding claim 17, it has been rejected in the same manner as claim 6.
Claims 2 and 13 are rejected under 35 U.S.C. § 103 as being unpatentable over Grossman in view of Hoon and further in view of Chandler et al. (U.S. 12,307,603 B2).
Regarding claim 2, Grossman as modified by Hoon, discloses the claimed invention, but fails to specifically disclose the one or more processors wherein the at least one neural network is updated using image pairs, at least one image pair depicting at least one feature of one or more objects in common and including data indicating a relative pose or position of a camera capturing each image of the at least one image pair.
In related art, Chandler discloses the one or more processors wherein the at least one neural network is updated using image pairs, at least one image pair depicting at least one feature of one or more objects in common and including data indicating a relative pose or position of a camera capturing each image of the at least one image pair. (Per Fig. 15A, Chandler discloses a camera pose related to each object. Chandler col. 34 lines 43–67. [a] camera pose relative to each object is estimated and refined in each frame based on the SDF model.)
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date to incorporate the teachings of Chandler into the teachings of Grossman and Hoon to provide annotated perception inputs fast and efficiently. Ibid. col. 4 lines 8–17.
Regarding claim 13, it has been rejected in the same manner as claim 2.
Allowable Subject Matter
Claims 3–4, 7–10, 14–15, 18 and 20 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.
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 BENEDICT LEE whose telephone number is (571)270-0390. The examiner can normally be reached 10:00-17:00 (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, Stephen R. Koziol can be reached at (408) 918-7630. 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.
/BENEDICT E LEE/Examiner, Art Unit 2665
/Stephen R Koziol/Supervisory Patent Examiner, Art Unit 2665
1 Applicant removed a condition view image in the amendment. This connotes that he leaves out the related mechanism because ¶53 discloses a relationship between the first and second sets of camera parameters. Without the condition view image, the specification does not provide validity of how the warping operation bridges the gap to a second set of camera parameters.
2 The specification may require a reasonable amount of experimentation to make and use the invention and what is reasonable will depend on the nature of the invention and the underlying art. Amgen Inc. et al. v. Sanofi et al., 598 U.S. 594, 596, 2023 USPQ2d 602 (2023).
3 See his col. 8 lines 34–48. Under a broadest reasonable interpretation (BRI), Examiner postulates that Grossmann uses current warping parameters in a stereo vision system to calculate corresponding pixel locations if the warping parameters are intact.
Prosecution Insights