CTNF 18/979,444 CTNF 100747 DETAILED ACTION Notice of Pre-AIA or AIA Status 07-03-aia AIA 15-10-aia The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA. Priority Applicant claims the benefit of US Provisional Application No. 63/573,203, filed 04/02/2024. Claims 1-20 have been afforded the benefit of this filing date. Information Disclosure Statement The IDS dated 06/24/2025 has been considered and placed in the application file. Claim Objections 07-29-01 AIA Claim s 1, 5, 11, and 20 objected to because of the following informalities: the words "frustum" and "frusta" are misspelled as "frustrum" (claim 5) and "frustra" (claims 1, 11, 20), respectively . Appropriate correction is required. Claim Rejections - 35 USC § 103 07-20-aia AIA 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. 07-21-aia AIA Claim (s) 1, 2, 5, 8-12, 15, 16, 19, and 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Huang et al. ("SelfOcc: Self-Supervised Vision-Based 3D Occupancy Prediction". arXiv preprint (29 Nov 2023). https://arxiv.org/abs/2311.12754v2; hereinafter "Huang"), in view of Philion et al. ("Lift, Splat, Shoot: Encoding Images From Arbitrary Camera Rigs by Implicitly Unprojecting to 3D". arXiv preprint (13 Aug 2020). https://arxiv.org/abs/2008.05711v1; hereinafter "Philion"), Yang et al. ("UniPAD: A Universal Pre-training Paradigm for Autonomous Driving". arXiv preprint (12 Oct 2023). https://arxiv.org/abs/2310.08370v1; hereinafter "Yang"), and Liu et al. ("Neural Sparse Voxel Fields". arXiv preprint (06 Jan 2021). https://arxiv.org/abs/2007.11571v2; hereinafter "Liu") . Regarding claim 1, Huang teaches: A computer-implemented method for generating generalized scene representations, the method comprising: extracting feature information from a plurality of scene images; encoding the feature information to generate a plurality of feature images (pg. 3 section 3.1 “From Image to Occupancy” describes supervised vision-based methods: “…a 2D backbone E first encodes N input images into multi-scale image features…”; the section goes on to describe how the invention of Huang uses a similar method; see fig. 3 “Image Backbone”) ; generating a plurality of voxels (pg. 2-3 section 3.1 “From Image to Occupancy”: “3D occupancy prediction is one prevalent proxy for scene reconstruction given its fine granularity and comprehensiveness, which aims at producing a voxelized prediction… encoding per-voxel occupancy (and semantic) information.”) ; sampling points along a plurality of views from different proposed camera angles relative to the plurality of octree voxels to produce feature angles and depths that are subsequently aggregated into a plurality of predicted feature maps (fig. 3 “To render a novel view, we apply a lightweight MLP on the 3D features to predict the SDF values, color and semantic vectors. We then perform volume rendering to synthesize color, depth and semantic views.”; pg. 3-4 section 3.2 “From Occupancy to Image”: “Take the volume rendering process of one single ray as an example. We first determine the origin and direction of the ray and uniformly sample M points P = {pm|m = 1, ..., M} between the origin and the intersection of the ray with the border of the 3D representation.”) ; and decoding the plurality of predicted feature maps to generate a plurality of final features maps (pg. 6 section 4.2 “Implementation Details”: “We further employ a two-layer MLP as the decoder network D to generate the SDF field, color and semantic logits (optional)”; fig. 3 shows color and semantic feature output) . Huang does not explicitly teach: estimating depths of at least a plurality of pixels in each feature image included in the plurality of feature images to produce a plurality of feature frustra; or generating a plurality of voxels from the plurality of feature frusta . Philion teaches: estimating depths of at least a plurality of pixels in each feature image included in the plurality of feature images to produce a plurality of feature frustra (pg. 2 “…our model “lifts” images into 3D by generating a frustum- shaped point cloud of contextual features”; fig. 3 shows depth estimation for individual pixels; section 3.1 “Lift: Latent Depth Distribution” explains this in detail) Huang and Philion are analogous to the claimed invention because they are in the same field of generating 3D scene representations for autonomous vehicles. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the invention of Huang with the teachings of Philion to convert 2D input images to 3D by estimating the depth for individual pixels. The motivation would have been substitution of one known element for another to obtain predictable results; both Huang and Philion include a method of converting 2D input images into a 3D scene representation, and the method taught by Philion (“Lift-Splat-Shoot”) is well known in the art and referenced by many other sources, including Huang itself. Huang teaches generating voxels from 2D information as previously discussed, but the combination of Huang in view of Philion does not explicitly teach: generating a plurality of voxels from the plurality of feature frusta . Yang teaches: generating a plurality of voxels from the plurality of feature frusta (pg. 3-4 section 3.2 “Unified 3D Volumetric Representation”: “For multi-view images, the view transformation [61] is adopted to transform 2D features into the 3D ego-car coordinate system to obtain the volume features. Specifically, we first predefine the 3D voxel coordinates…, where X × Y × Z is the voxel resolution. Subsequently, the Xp is projected on multi-view images to index the corresponding 2D features…”; pg. 2 col. 1 teaches that the 3D volume is constructed using the method from Philion, which generates feature frustra from input images). Yang is analogous to the claimed invention because it is in the same field of generating 3D scene representations for autonomous vehicles. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the invention of Huang in view of Philion with the teachings of Yang to convert a 3D frustum created from an input image to a voxel representation in a reference coordinate frame. The motivation would have been substitution of one known element for another to obtain predictable results; both Huang and Yang include a method of converting input images into 3D voxels, and the invention of Yang is explicitly designed to work with the method of Philion in which 2D feature images are converted to 3D feature frustra. The combination of Huang in view of Philion and Yang does not explicitly teach: generating a plurality of octree voxels. Liu teaches representing voxels using an octree data structure (pg. 3 section 3 “Neural Sparse Voxel Fields”: “Instead of representing the entire scene as a single implicit field, NSVF consists of a set of voxel-bounded implicit fields organized in a sparse voxel octree.”) Liu is analogous to the claimed invention because it is in the same field of neural rendering for novel viewpoint synthesis. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the invention of Huang in view of Philion and Yang with the teachings of Liu to represent the voxels using an octree. The motivation would have been to improve the speed of novel viewpoint synthesis (the method of Liu was 10x faster than the state of the art at the time – see the abstract). Regarding claim 2, the combination of Huang in view of Philion, Yang, and Liu teaches: The computer-implemented method of claim 1, wherein the plurality of scene images comprises a set of images captured by one or more vehicle cameras (Huang fig. 1 shows input images are taken from 6 vehicle cameras at different angles) . Regarding claim 5, the combination of Huang in view of Philion, Yang, and Liu teaches: The computer-implemented method of claim 1, wherein generating the plurality of octree voxels comprises combining one of more features included in each feature frustrum into a feature volume (Philion fig. 4 “In the “lift” step, a frustum-shaped point cloud is generated for each individual image (center-left)”; pg. 5 section 3.1 “Lift: Latent Depth Distribution”: “The purpose of this stage is to “lift” each image from a local 2-dimensional coordinate system to a 3-dimensional frame that is shared across all cameras.”) , and performing at least one of one or more quantization or one or more convolution operations on the feature value to produce a series of octrees (Yang pg. 4 section 3.2 “Unified 3D Volumetric Representation” discusses how the voxel resolution is determined and the voxel grid is projected onto 2D features, which can be considered “quantization”; Liu discusses storing voxels in an octree structure as previously discussed for claim 1) . The motivation to combine the teachings of these inventions would have been the same as for claim 1. Regarding claim 8, the combination of Huang in view of Philion, Yang, and Liu teaches: The computer-implemented method of claim 1, wherein a different predicted feature map is produced for each proposed camera angle (Huang fig. 3: “To render a novel view, we apply a lightweight MLP on the 3D features to predict the SDF values, color and semantic vectors. We then perform volume rendering to synthesize color, depth and semantic views.”; this is done for each novel view) . Regarding claim 9, the combination of Huang in view of Philion, Yang, and Liu teaches: The computer-implemented method of claim 1, wherein decoding the plurality of predicted feature maps comprises applying one or more supplemental transformations to the plurality of predicted feature maps to generate the plurality of final feature maps (Huang fig. 3 shows semantic segmentation map applied to rendered output image: “To render a novel view, we apply a lightweight MLP on the 3D features to predict the SDF values, color and semantic vectors.”) . Regarding claim 10, the combination of Huang in view of Philion, Yang, and Liu teaches: The computer-implemented method of claim 9, wherein the one or more supplemental transformations include a first transformation, and wherein the first transformation is applied to every predicted feature map included in the plurality of predicted feature maps (Huang fig. 3 shows semantic segmentation map applied to rendered output image: “To render a novel view, we apply a lightweight MLP on the 3D features to predict the SDF values, color and semantic vectors. We then perform volume rendering to synthesize color, depth and semantic views”; the caption suggests that the generation of a semantic segmentation map this is done for each novel view) . Regarding claim 11, it is rejected using the same references, rationale, and motivation to combine as claim 1 because its limitations substantially correspond to the limitations of claim 1, as well as the additional limitation of: One or more non-transitory, computer-readable media storing instructions that, when executed by one or more processors, cause the one or more processors to perform the steps… (Huang pg. 9-10 “Training settings” specifies the use of standard PC-compatible datasets and hardware, including the use of 8 RTX-3090 GPUs with 24GB memory; one of ordinary skill in the art may infer that the invention is implemented as a standard computer program saved on a standard PC-compatible storage device). Regarding claim 12, it is rejected using the same references, rationale, and motivations to combine as claim 2 because its limitations substantially correspond to the limitations of claim 2. Regarding claim 15, the combination of Huang in view of Philion, Yang, and Liu teaches: The one or more non-transitory, computer-readable media of claim 11, wherein a decoder module performs at least one of object detection or classification on the plurality of predicted feature maps (Huang pg. 6 section 4.2 “Implementation Details”: “We further employ a two-layer MLP as the decoder network D to generate the SDF field, color and semantic logits (optional)”; fig. 3 shows semantic segmentation output) . Regarding claim 16, the combination of Huang in view of Philion, Yang, and Liu teaches: The one or more non-transitory, computer-readable media of claim 11, wherein the steps of extracting feature information, encoding the feature information, estimating the depths of at least a plurality of pixels, generating the plurality of octree voxels, sampling points along the plurality of views, and decoding the plurality of predicted feature maps are performed by a scene representation prediction application, and wherein the scene representation prediction application is trained using training data generated using a plurality of neural radiance fields (Huang pg. 2 “Self-supervised Depth Prediction” section teaches the training of NeRFs: “…we propose a novel MVS-embedded approach to learn depth in NeRFs.”; pg. 4-5 section 3.3 “Occupancy-Oriented Supervision” explains further; table 4 shows the results of training using only monocular input images, where corresponding depth values were obtained via NeRF model) . Regarding claim 19, it is rejected using the same references, rationale, and motivations to combine as claim 9 because its limitations substantially correspond to the limitations of claim 9. Regarding claim 20, it is rejected using the same references, rationale, and motivation to combine as claim 1 because its limitations substantially correspond to the limitations of claim 1, as well as the additional limitation of: A computer system, comprising: one or more memories storing instructions; and one or more processors that are coupled to the one or more memories and, when executing the instructions, are configured to perform the steps… (Huang pg. 9-10 “Training settings” specifies the use of standard PC-compatible datasets and hardware, including the use of 8 RTX-3090 GPUs with 24GB memory, each of which contains a processor; one of ordinary skill in the art may infer that the invention is implemented as a standard computer program which may be run using the GPUs) . 07-22-aia AIA Claim (s) 3, 4, and 13 is/are rejected under 35 U.S.C. 103 as being unpatentable over Huang ("SelfOcc: Self-Supervised Vision-Based 3D Occupancy Prediction) in view of Philion ("Lift, Splat, Shoot: Encoding Images From Arbitrary Camera Rigs by Implicitly Unprojecting to 3D"), Yang ("UniPAD: A Universal Pre-training Paradigm for Autonomous Driving"), and Liu ("Neural Sparse Voxel Fields") as applied to claim s 1, 11, and 16 above, and further in view of Ye et al. ("FeatureNeRF: Learning Generalizable NeRFs by Distilling Foundation Models". arXiv preprint (22 Mar 2023). https://arxiv.org/abs/2303.12786v1; hereinafter "Ye") . Regarding claim 3, the combination of Huang in view of Philion, Yang, and Liu teaches: The computer-implemented method of claim 1, but does not explicitly teach: wherein the feature information comprises object information inferred from a scene by a foundation model . Ye teaches: wherein the feature information comprises object information inferred from a scene by a foundation model (fig. 2 “we distill knowledge from 2D vision foundation models to FeatureNeRF.”; Section 3.2 “Feature Distillation from Foundation Models” equation (4); Section 3.4 “Applications of FeatureNeRF”: “Given a single image I , we can render its NeRF feature map… using Eq. 4.”) . Ye is analogous to the claimed invention because it is in the same field of 3D scene reconstruction using neural radiance fields. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the invention of Huang in view of Philion, Yang, and Liu with the teachings of Ye to learn feature information using a foundation model. The motivation would have been to better learn semantic information, enabling tasks requiring semantic information such as segmentation (Ye pg. 2 col. 1). Regarding claim 4, the combination of Huang in view of Philion, Yang, and Liu and further in view of Ye teaches: The computer-implemented method of claim 3, wherein the foundation model encodes the object information to generate the plurality of feature images (Ye fig. 2 encoder; pg. 2 col. 1 “Specifically, we adopt an encoder to map 2D images to corresponding 3D NeRF volume similar to previous generalizable NeRFs. Apart from density and color, we propose to extract deep features of the query 3D points from the intermediate layers of NeRF MLP. To enrich semantic information of the NeRF features, we further transfer knowledge from the foundation models to the encoder via neural rendering during training”) . The motivation to modify the invention of Huang in view of Philion, Yang, and Liu and further in view of Ye with the additional teachings of Ye would have been the same as for claim 3. Regarding claim 13, it is rejected using the same references, rationale, and motivations to combine as claim 3 because its limitations substantially correspond to the limitations of claim 3 . 07-22-aia AIA Claim (s) 6 is/are rejected under 35 U.S.C. 103 as being unpatentable over Huang ("SelfOcc: Self-Supervised Vision-Based 3D Occupancy Prediction) in view of Philion ("Lift, Splat, Shoot: Encoding Images From Arbitrary Camera Rigs by Implicitly Unprojecting to 3D"), Yang ("UniPAD: A Universal Pre-training Paradigm for Autonomous Driving"), and Liu ("Neural Sparse Voxel Fields") as applied to claim 1 above, and further in view of Irshad et al. (US 20240028792 A1; hereinafter "Irshad") . Regarding claim 6, the combination of Huang in view of Philion, Yang, and Liu teaches: The computer-implemented method of claim 5, but does not explicitly teach: wherein the octrees included in the series of octrees have differing resolutions. Irshad teaches: wherein the octrees included in the series of octrees have differing resolutions ([0081] “The refinement uses an Octree-based coarse-to-fine differentiable optimization to improve shape, appearance, pose, and size predictions iteratively.”; paragraphs [0089] and [0090] explain how octree voxels may be sampled at different levels of detail) . Irshad is analogous to the claimed invention because it is in the same field of neural volume rendering. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the invention of Huang in view of Philion, Yang, and Liu to incorporate the octree-based coarse-to-fine differentiable optimization of Irshad. The motivation would have been to improve memory and computational efficiency (Irshad [0089]) . 07-22-aia AIA Claim (s) 7 and 18 is/are rejected under 35 U.S.C. 103 as being unpatentable over Huang ("SelfOcc: Self-Supervised Vision-Based 3D Occupancy Prediction) in view of Philion ("Lift, Splat, Shoot: Encoding Images From Arbitrary Camera Rigs by Implicitly Unprojecting to 3D"), Yang ("UniPAD: A Universal Pre-training Paradigm for Autonomous Driving"), and Liu ("Neural Sparse Voxel Fields") as applied to claim s 1 and 11 above, and further in view of Hasan et al. (US 20250139883 A1; hereinafter "Hasan") . Regarding claim 7, the combination of Huang in view of Philion, Yang, and Liu teaches: The computer-implemented method of claim 1, wherein the feature angles and depths are subsequently aggregated into the plurality of predicted feature maps via a ray marching procedure applied to a plurality of importance-sampled points. The combination of Huang in view of Philion, Yang, and Liu does not explicitly teach that the points are importance-sampled . Hasan teaches importance sampling when performing volume rendering (fig. 4; [0042] “Rendering component 220 performs a ray marching process to identify locations for color sampling by color sampling component 215, and then aggregates the color information and density information along locations of a pixel ray to yield a final color for the corresponding pixel. The number of samples and the locations of the samples are determined according to an importance sampling method , which will be described in detail with reference to FIG. 4.”). Hasan is analogous to the claimed invention because it is in the same field of 3D scene reconstruction and novel view synthesis using neural radiance fields. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the invention of Huang in view of Philion, Yang, and Liu with the teachings of Hasan to sample color information along a ray according to an importance sampling method. The motivation would have been that it “greatly reduces the number of samples used to produce pixel images” (Hasan [0019]), therefore improving performance. Regarding claim 18, it is rejected using the same references, rationale, and motivations to combine as claim 7 because its limitations substantially correspond to the limitations of claim 7 . 07-22-aia AIA Claim (s) 14 is/are rejected under 35 U.S.C. 103 as being unpatentable over Huang ("SelfOcc: Self-Supervised Vision-Based 3D Occupancy Prediction) in view of Philion ("Lift, Splat, Shoot: Encoding Images From Arbitrary Camera Rigs by Implicitly Unprojecting to 3D"), Yang ("UniPAD: A Universal Pre-training Paradigm for Autonomous Driving"), and Liu ("Neural Sparse Voxel Fields") as applied to claim 11 above, and further in view of Su et al. (US 20250308142 A1; hereinafter "Su") . Regarding claim 14, the combination of Huang in view of Philion, Yang, and Liu teaches: The one or more non-transitory, computer-readable media of claim 11, but does not explicitly teach: wherein decoding the plurality of predicted feature maps comprises enhancing high-frequency details included in at least one predicted feature map or increasing a resolution associated with at least one predicted feature map. Su teaches: wherein decoding the plurality of predicted feature maps comprises enhancing high-frequency details included in at least one predicted feature map or increasing a resolution associated with at least one predicted feature map ([0039] “Spatial resolution: using an upscaled version of the base layer, one can add the enhancement layer information to enhance the details of a final scene at a higher resolution than the base layer resolution”) . Su is analogous to the claimed invention because it is in the same field of 3D scene reconstruction and novel view synthesis using neural radiance fields. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the invention of Huang in view of Philion, Yang, and Liu with the teachings of Su to perform upscaling to increase the resolution of a NeRF output. The motivation would have been to “enhance the details of a final scene” as taught by Su . 07-22-aia AIA Claim (s) 17 is/are rejected under 35 U.S.C. 103 as being unpatentable over Huang ("SelfOcc: Self-Supervised Vision-Based 3D Occupancy Prediction) in view of Philion ("Lift, Splat, Shoot: Encoding Images From Arbitrary Camera Rigs by Implicitly Unprojecting to 3D"), Yang ("UniPAD: A Universal Pre-training Paradigm for Autonomous Driving"), and Liu ("Neural Sparse Voxel Fields") as applied to claim 16 above, and further in view of Ye ("FeatureNeRF: Learning Generalizable NeRFs by Distilling Foundation Models") and Valentin et al. (US 12182922 B2; hereinafter "Valentin") . Regarding claim 17, the combination of Huang in view of Philion, Yang, and Liu teaches: The one or more non-transitory, computer-readable media of claim 16, wherein the plurality of neural radiance fields are used to generate depth estimates for training scene images (Huang pg. 2 col. 1 “our method can reconstruct meaningful 3D occupancy in a self-supervised manner”; where “self-supervised” means that the depth is predicted from the training images rather than provided from additional 3D data such as LiDAR). The combination of Huang in view of Philion, Yang, and Liu does not explicitly teach: and wherein a foundation model transforms the full set of training images and depths into the training data. Ye teaches: wherein a foundation model transforms the full set of training images and depths into the training data (pg. 2 col. 1 “Specifically, we adopt an encoder to map 2D images to corresponding 3D NeRF volume similar to previous generalizable NeRFs. Apart from density and color, we propose to extract deep features of the query 3D points from the intermediate layers of NeRF MLP. To enrich semantic information of the NeRF features, we further transfer knowledge from the foundation models to the encoder via neural rendering during training: The rendered feature outputs should be consistent with the feature extracted from the foundation models, which is enforced by a distillation loss.”) . The motivation to modify the invention of Huang in view of Philion, Yang, and Liu and further in view of Ye with the additional teachings of Ye would have been the same as for claim 3. The combination of Huang in view of Philion, Yang, and Liu and further in view of Ye does not explicitly teach: the training scene images and the depth estimates are combined with synthetic images and depth estimates into a full set of training images and depths . Valentin teaches the use of synthetic images for NeRF training (col. 13 lines 49-51 “The training data images are real images such as photographs or video frames. It is also possible for the training data images to be synthetic images.”). Valentin is analogous to the claimed invention because it is in the same field of 3D neural rendering. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the invention of Huang in view of Philion, Yang, and Liu with the teachings of Valentin to augment the training data using synthetic images. The motivation would have been to increase the quantity of training data to produce better results. References Cited 07-96 AIA The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. Wang et al. "SparseNeRF: Distilling Depth Ranking for Few-shot Novel View Synthesis". arXiv preprint (13 Aug 2023). https://arxiv.org/abs/2303.16196v2 Conclusion Any inquiry concerning this communication or earlier communications from the examiner should be directed to BENJAMIN STATZ whose telephone number is (571)272-6654. The examiner can normally be reached Mon-Fri 8am-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, Tammy Goddard can be reached at (571)272-7773. 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. /BENJAMIN TOM STATZ/ Examiner, Art Unit 2611 /TAMMY GODDARD/ Supervisory Patent Examiner, Art Unit 2611 Application/Control Number: 18/979,444 Page 2 Art Unit: 2611 Application/Control Number: 18/979,444 Page 3 Art Unit: 2611 Application/Control Number: 18/979,444 Page 4 Art Unit: 2611 Application/Control Number: 18/979,444 Page 5 Art Unit: 2611 Application/Control Number: 18/979,444 Page 6 Art Unit: 2611 Application/Control Number: 18/979,444 Page 7 Art Unit: 2611 Application/Control Number: 18/979,444 Page 8 Art Unit: 2611 Application/Control Number: 18/979,444 Page 9 Art Unit: 2611 Application/Control Number: 18/979,444 Page 10 Art Unit: 2611 Application/Control Number: 18/979,444 Page 11 Art Unit: 2611 Application/Control Number: 18/979,444 Page 12 Art Unit: 2611 Application/Control Number: 18/979,444 Page 13 Art Unit: 2611 Application/Control Number: 18/979,444 Page 14 Art Unit: 2611 Application/Control Number: 18/979,444 Page 15 Art Unit: 2611 Application/Control Number: 18/979,444 Page 16 Art Unit: 2611 Application/Control Number: 18/979,444 Page 17 Art Unit: 2611 Application/Control Number: 18/979,444 Page 18 Art Unit: 2611 Application/Control Number: 18/979,444 Page 19 Art Unit: 2611
Prosecution Insights