DETAILED ACTION
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 .
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) 1-3, 6-7, 10-14 and 17-18 is/are rejected under 35 U.S.C. 103 as being unpatentable over Zakharov et al. (U.S. Patent Application Publication No. 2022/0414974), hereinafter referenced as Zakharov, in view of Menapace et al. (U.S. Patent Application Publication No. 2024/0307783), hereinafter referenced as Menapace, and Gupta et al. (3DGen: Triplane Latent Diffusion for Textured Mesh Generation), hereinafter referenced as Gupta.
Regarding claim 1, an estimation system comprising: (fig. 2 teaches system 140 and abstract teaches “systems…estimates”); a memory storing instructions that, when executed by a processor, cause the processor to: (fig. 2 teaches memory 210 and processor 205 and paragraph 4 teaches “memory also stores … instructions that when executed by the one or more processors cause the one or more processors to”); estimate a normalized object reference frame (NORF) image (paragraph 4 teaches “a Normalized Object Coordinate Space (NOCS) map of the object” and paragraph 43 teaches “Given estimated object masks and NOCS maps…(2) 3D points of the partial 3D object shape 255 in the canonical frame of reference,”); canonical frame of reference would be reference frame and having NOCS estimated indicates a NORF; and a NORF normal for an object from an image (paragraph 48 teaches “associated surface normals representing rendered objects”); surface normal indicates object in NORF would have a normal; the object having incomplete data (paragraph 43 teaches “3D points of the partial 3D object shape 255”); partial object shape shows object having incomplete data; derive a projection of the object from a point cloud using the NORF image and the NORF normal (paragraph 44 teaches “In one embodiment, a 0-isosurface projection (marked “P” in the figure) is used to retrieve an object's surface from an SDF field” and paragraph 48 teaches “to train the SDF module, a collection of canonical ground-truth point clouds and associated surface normals representing rendered objects are used”); projection P shows projection of object and since from SDF field which is trained on canonical ground-truth point clouds and associated surface normal, this is from a point cloud using the NORF image and NORF normal; and predict a completed shape for the object from the projection (paragraph 50 teaches “initially predicted SDF is used to recover the full surface of the object in the form of a point cloud with surface normal using a 0-isosurface projection”); full surface shows completed shape for object and is predicted since uses initially predicted SDF, also, it is from the projection since using projection.
However, Zakharov fails to explicitly teach estimate…and noise using a NORF diffusion model;
However, Menapace teaches estimate…and noise using a NORF diffusion model (Menapace, paragraph 42 teaches the “diffusion-based animation model 120 predicts noise ϵ.sub.k applied to the noisy states s.sup.p.sub.k conditioned on known partial states… Each object is represented in its canonical pose”); this shows noise estimated/predicted using a diffusion model which is considered NORF diffusion model when viewed in combination since it also has canonical / normalized object information as above. Menapace is considered to be analogous art because it is reasonably pertinent to the problem faced by the inventor of noise prediction using diffusion model for image processing and graphics rendering. Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Zakharov's invention with the noise prediction and diffusion model techniques of Menapace to improve realism (Menapace, paragraph 36). This would be due to the use of the diffusion model for better quality.
However, the combination of Zakharov and Menapace fails to explicitly teach predict… and triplanar noise using a triplanar diffusion model.
However, Gupta teaches predict… and triplanar noise using a triplanar diffusion model (Gupta, page 3, section 2.4, first paragraph teaches “After obtaining a VAE with a well trained triplane latent space, we train a diffusion model to generate these features… We train this model with 1000 denoising steps, use the v prediction objective and a cosine noise schedule”); using prediction objective and cosine noise for training shows predicting triplanar noise and since the well trained latent space is generated using trained diffusion model, it indicates a triplanar diffusion model. Gupta is considered to be analogous art because it is reasonably pertinent to the problem faced by the inventor of triplanar noise and diffusion model thereof for image generation. Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify the combination of Zakharov and Menapace with the triplanar techniques of Gupta to improve 3D consistency and overall quality (Gupta, page 3, section 2.4, first paragraph). This improved 3D shape and consistency would be due to the three planes and improved coherency from such.
Regarding claim 2, the combination of Zakharov, Menapace and Gupta teaches further including instructions to: register a lifted representation of the object outputted by the NORF diffusion model, (Zakharov, paragraph 43 teaches “2D coordinates of the object in the image 305, (2) 3D points of the partial 3D object shape 255 in the canonical frame of reference, and (3) their 2D-3D correspondences”); the 2D to 3D shows lifting between the dimensions and when viewed in combination this would be the aforementioned object outputted by the NORF diffusion model of Menapace; the lifted representation associated with the point cloud that is incomplete about the object (Zakharov, paragraph 43 teaches pose is then estimated using a PnP algorithm that predicts the object pose (260) from given correspondences”); PnP algorithm predicting means point cloud is being generated (thus currently incomplete) also since the aforementioned 3D points are of partial 3D object; and estimate a metric pose of the object within a scene using inputted depth (Zakharov, paragraph 43 teaches “multiplied by the per-class scale factor to recover absolute scale. The six-degrees-of-freedom (6DoF) pose is then estimated” and paragraph 42 teaches “This architecture takes masked RGB and depth images and inpaints RGB and depth values for occluded regions, employing a fully convolutional PatchGAN discriminator to judge the genuineness of the inpainted maps”); metric pose is shown by multiplying by scale factor to recover absolute scale and 6DOF all of which is done using the depth values for occluded regions since comes in a step after ; and one of multiple hypotheses about the projection associated with the lifted representation, (Zakharov, paragraph 43 teaches “Since there are a large set of correspondences for each model, PnP is, in some embodiments, combined with RANSAC to increase the robustness of the system against outliers”); RANSAC uses one of a multiple hypotheses by fitting a model (which is done by hypothesizing the right fit); the inputted depth is part of the image and the image represents a single view (Zakharov, paragraph 18 teaches “performs 3D scene reconstruction from a single 2D image” and paragraph 4 teaches “and the depth map for the background portion of the scene”); this shows image representing single view and the aforementioned depth would be part of that image. The same motivations used in claim 1 apply here in claim 2.
Regarding claim 3, the combination of Zakharov, Menapace and Gupta teaches further including instructions to: map the image by the NORF diffusion model to a reference frame, (Menapace, paragraph 70 teaches “mesh is aligned to its desired position in the object's canonical space”); object in canonical pose would show image by the NORF diffusion model and it is mapped/aligned to mesh which acts as a reference frame; wherein the image is segmented (Menapace, paragraph 29 teaches “incorporate various conditioning modalities in addition to text, such as segmentation masks or partially occluded images”); segmentation mask shows image as segmented; sample the image within the reference frame using the NORF diffusion model (Menapace, paragraph 42 teaches “the animation model 120 outputs sampled properties of the objects,”); this uses the aforementioned NORF diffusion model (animation model) and samples image within reference frame since refers to the sampled properties of object and object would come from such frame; generate the point cloud that is incomplete by lifting the NORF image from two-dimensions to three-dimensions using the image and the NORF normal, (Zakharov, paragraph 43 teaches “2D coordinates of the object in the image 305, (2) 3D points of the partial 3D object shape 255 in the canonical frame of reference, and (3) their 2D-3D correspondences… recovered normalized shape is multiplied by the per-class scale factor to recover absolute scale… pose is then estimated using a PnP algorithm that predicts the object pose (260) from given correspondences”); PnP algorithm predicting means point cloud is being generated (thus currently incomplete also since the aforementioned 3D points are of partial 3D object), the 2D to 3D shows lifting between the dimensions and this is using NORF normal since uses normalized shape as well as the aforementioned image; wherein the point cloud is associated with a correspondence between pixels of the image and three-dimensional (3D) coordinate points (Zakharov, paragraph 43 teaches “(1) 2D coordinates of the object in the image,,, and (3) their 2D-3D correspondences”); this shows the aforementioned point cloud is associated with correspondence between pixels of image (2D coordinates) and the corresponding 3D coordinate points; and the projection includes information from the point cloud (Zakharov, paragraph 51 teaches “estimating surface points and normal vector using a 0-isosurface projection”); since surface points use projection, the projection includes information from point cloud; and position the object in an actual scene using the 3D coordinate points (Zakharov, paragraph 61 teaches “block 660, 3D reasoning module 225 generates an editable and re-renderable 3D reconstruction of the scene 270 based, at least in part, on the complete shape of the object, the refined estimated pose of the object 265, and the depth map… translating the object within the editable and re-renderable 3D reconstruction of the scene”); this shows positioning the object in actual 3D scene since the scene is based on object and translating shows it would be done using 3D coordinate points so that object can be accurately/precisely moved later. The same motivations used in claim 1 apply here in claim 3.
Regarding claim 6, the combination of Zakharov, Menapace and Gupta teaches wherein the completed shape includes a geometry of the object (Zakharov, paragraph 41 teaches “This partial 3D shape 255 enables the processing pipeline to retrieve the object's full geometry and pose”); this shows completed shape including geometry.
Regarding claim 7, the combination of Zakharov, Menapace and Gupta
wherein: the NORF image is associated with a NORF position map having pixel colors representing different 3D positions in a reference frame (Zakharov, paragraph 41 teaches “foreground is predicted as a set of normalized shape maps (NOCS maps 240 produced by a NOCS network 345). Since they encode 3D coordinates via RGB, visualizing each 3D color component in 3D space”); 3D coordinates encoded via RGB shows pixel colors representing different 3D positions/coordinates, this is for NOCS map which acts as NORF position map (when it has positions encoded and is for the aforementioned canonical reference frame) and NOCS map here is associated with the NORF image of claim 1; and the NORF image is associated with NORF normal map having a pixel value representing a surface normal of the object from an observed point (Zakharov, paragraph 48 teaches “collection of canonical ground-truth point clouds and associated surface normals representing rendered objects” and paragraph 50 teaches “initially predicted SDF is used to recover the full surface of the object in the form of a point cloud with surface normal using a 0-isosurface projection. Then, 2D/3D optimization process 370 estimates nearest neighbors between the two point clouds and minimizes the distance between the points”); canonical point cloud having associated surface normals representing rendered object and full surface of object recovered from such shows the NORF image is associated with NORF normal map (since surface normal associated with surface would have mapping for the association) and pixel values in such must be present to represent the surface normal of the object which would be from an observed viewpoint.
Regarding claim 10, the non-transitory computer-readable medium claim 10 recites similar limitations as system claim 1, and thus is rejected under similar rationale. In addition, Zakharov, paragraph 5 teaches “Another embodiment is a non-transitory computer-readable medium”.
Regarding claim 11, the non-transitory computer-readable medium claim 11 recites similar limitations as system claim 2, and thus is rejected under similar rationale.
Regarding claim 12, the method claim 12 recites similar limitations as system claim 1, and thus is rejected under similar rationale.
Regarding claim 13, the method claim 13 recites similar limitations as system claim 2, and thus is rejected under similar rationale.
Regarding claim 14, the method claim 14 recites similar limitations as system claim 3, and thus is rejected under similar rationale.
Regarding claim 17, the method claim 17 recites similar limitations as system claim 6, and thus is rejected under similar rationale.
Regarding claim 18, the method claim 18 recites similar limitations as system claim 7, and thus is rejected under similar rationale.
Claim(s) 4 and 15 is/are rejected under 35 U.S.C. 103 as being unpatentable over the combination of Zakharov, Menapace and Gupta as applied to claim 1 above, and further in view of Gudivada et al. (Face recognition using ortho-diffusion bases), hereinafter referenced as Gudivada.
Regarding claim 4, the combination of Zakharov, Menapace and Gupta teaches wherein the instructions to predict the completed shape further include instructions to: data derived from the projection using the triplanar noise by the triplanar diffusion model (Gupta, page 3, section 2.4, first paragraph teaches “After obtaining a VAE with a well trained triplane latent space, we train a diffusion model to generate these features… We train this model with 1000 denoising steps, use the v prediction objective and a cosine noise schedule”); predicting complete shape using data derived from projection is shown in Zakharov as mentioned in claim 1 above, cosine noise for training shows using triplanar noise and since the well trained latent space is generated using trained diffusion model, it indicates a triplanar diffusion model for performing the following mentioned and diffusing tasks; and extract the object from a triplanar representation by the triplanar diffusion model (Gupta, page 2, fig. 1 teaches textured mesh (extracted object) resulting from a triplanar representation by the triplanar diffusion model).
However, the combination of Zakharov, Menapace and Gupta fails to teach diffuse ortho-normal data; wherein the ortho-normal data is a condition
However, Gudivada teaches diffuse ortho-normal data (Gudivada, abstract teaches “At each recursion a set of orthonormal bases functions are extracted for a specific scale. A diffusion step is embedded at each scale in the QR decomposition”); this shows diffusion of ortho-normal data; wherein the ortho-normal data is a condition (Gudivada, page 2, first full paragraph teaches “In the following we describe how to use the orthogonal decompositions of diffusion wavelets in order to extract the ortho-diffusion faces from the matrix C”); this shows orthonormal data as a condition since it is used for decomposition and extraction. Gudivada is considered to be analogous art because it is reasonably pertinent to the problem faced by the inventor of ortho-normal data. Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify the combination of Zakharov, Menapace and Gupta with the ortho-normal techniques of Gudivada to improve the recognition results (Gudivada, page 4, second paragraph). This would lead to a more accurately rendered and reconstructed full object as well.
Regarding claim 15, the method claim 15 recites similar limitations as system claim 4, and thus is rejected under similar rationale.
Claim(s) 5 and 16 is/are rejected under 35 U.S.C. 103 as being unpatentable over the combination of Zakharov, Menapace and Gupta as applied to claim 1 above, and further in view of Xu et al. (U.S. Patent Application Publication No. 2024/0265621), hereinafter referenced as Xu.
Regarding claim 5, the combination of Zakharov, Menapace and Gupta teaches wherein the instructions to predict the completed shape further include instructions to: using the projection of the point cloud having partial information, (Zakharov, paragraph 51 teaches “estimating surface points and normal vector using a 0-isosurface projection” and paragraph 43 teaches pose is then estimated using a PnP algorithm that predicts the object pose (260) from given correspondences”); PnP algorithm predicting means point cloud is being generated (thus currently incomplete also since the aforementioned 3D points are of partial 3D object) and since surface points use projection, the projection is of the point cloud; and the triplanar neural field represents a prior of object shapes (Zakharov, paragraph 4 teaches “multilayer perceptrons (MLPs) to produce a second latent vector for the object that represents the object in a differentiable database of object priors. The differentiable database of object priors encodes the geometry of the object priors”); this shows prior of object shapes used in the process of predicting completed shape and when viewed in combination it would be from the triplanar neural field of the below reference; and the triplanar neural field including signed distance fields (Zakharov, paragraph 4 teaches “using signed distance fields (SDFs)”); when viewed in combination the SDF would be included in the triplanar neural field of the below reference.
However, the combination of Zakharov, Menapace and Gupta fails to explicitly teach represent the object as a triplanar neural field.
However, Xu teaches represent the object as a triplanar neural field (Xu, paragraph 57 teaches “To synthesize images out of training distribution of expressions (β, θ), triplane generation may be disentangled but when decoding neural feature field (σ, f) from sampled triplane features, the MLP may also be conditioned on β and θ”); this shows the object represented as a triplanar neural field since neural feature field when decoding has triplane features. Xu is considered to be analogous art because it is reasonably pertinent to the problem faced by the inventor of objects as triplanar neural field. Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify the combination of Zakharov, Menapace and Gupta with the triplanar neural field techniques of Xu to improve control accuracy of the output synthesized images (Xu, paragraph 49). This would further improve the realism due to improved accuracy.
Regarding claim 16, the method claim 16 recites similar limitations as system claim 5, and thus is rejected under similar rationale.
Claim(s) 8 and 19 is/are rejected under 35 U.S.C. 103 as being unpatentable over the combination of Zakharov, Menapace and Gupta as applied to claim 1 above, and further in view of Brion et al. (U.S. Patent Application Publication No. 2023/0113698), hereinafter referenced as Brion.
Regarding claim 8, the combination of Zakharov, Menapace and Gupta teaches wherein: the NORF diffusion model and the triplanar diffusion model are a diffusion denoising probabilistic model (Menapace, paragraph 82 teaches “Diffusion models have recently shown state-of-the-art performance on several tasks such as text-conditioned image and video generation, sequence modeling, and text-conditioned human motion generation. Thus, the animation model 120 may be based on the denoising diffusion probabilistic models (DDPM) diffusion framework”); this shows the NORF diffusion model being a diffusion denoising probabilistic model (DDPM), and one of ordinary skill in the art would understand that the same logic can be applied to the triplanar diffusion model from Gupta to make it a DDPM so that state-of the-art performance can be reached there as well; that individually output multiple hypotheses about one of the projection and the completed shape, (Zakharov, paragraph 36 teaches “reconstructs the scene by not only explaining every visible pixel but also predicting the full geometry” and paragraph 39 teaches “predict important properties for each detected object. In one embodiment, a 3-layer perceptron with ReLU activation and a linear projection layer is used”); predicting here shows hypotheses and is for the completed/full shape as well as projection (due to mention of projection layer used for predicting important properties); and the projection includes partial shape and pose information (Zakharov, paragraph 51 teaches “Differentiable rendering allows 2D/3D optimization process 370 to optimize objects with respect to both pose and shape… renderer that uses surfels as the primary representation, estimating surface points and normal vector using a 0-isosurface projection followed by a surfel-based rasterizer”); surfels show projection including partial shape, and optimizing with respect to both pose and shape shows the projection would include both pose and shape as well;
However, the combination of Zakharov, Menapace and Gupta fails to teach
and the projection has orthogonal NORF data from voxelizing and tracing three-dimensional points of the point cloud that are incomplete.
However, Brion teaches
and the projection has orthogonal NORF data from voxelizing and tracing three-dimensional points of the point cloud that are incomplete (Brion, claim 11 teaches “skeletonizing strands of a composite material part in a 3D image of said part, said image comprising a plurality of voxels… said voxels extending in the orthogonal coordinates system”); this shows the projection from above combination (with NORF data) would have orthogonal data which is from voxelizing (since part of image here comprises voxels) and tracing (since voxels extend in orthogonal coordinate system thus the points inside the voxel would too) the aforementioned incomplete point cloud’s 3D points (as explained above in this action and from Zakharov, paragraph 43). Brion is considered to be analogous art because it is reasonably pertinent to the problem faced by the inventor of orthogonal data and voxelizing. Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify the combination of Zakharov, Menapace and Gupta with the orthogonal and voxel techniques of Brion to check that the centers are not too far from one another, which could indicate a potential strand defect, such as a discontinuity (Brion, paragraph 37). This would ensure error handling leading to a better user experience since defects in the object formation could be caught beforehand when using these voxelizing and orthogonal techniques.
Regarding claim 19, the method claim 19 recites similar limitations as system claim 8, and thus is rejected under similar rationale.
Claim(s) 9 and 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over the combination of Zakharov, Menapace and Gupta as applied to claim 1 above, and further in view of Brumby et al. (U.S. Patent Application Publication No. 2024/0362898), hereinafter referenced as Brumby, and Gudivada.
Regarding claim 9, the combination of Zakharov, Menapace and Gupta teaches wherein: the NORF diffusion model is associated with a first probabilistic distribution of the object within a reference frame in a first stage (Menapace, paragraph 82 teaches “animation model 120 may be based on the denoising diffusion probabilistic models (DDPM) diffusion framework”); since animation/NORF diffusion model is based on diffusion probabilistic models, a first probabilistic distribution of object within reference frame in at least a first stage would be present; the NORF diffusion model is conditioned with the image and the noise that is two- dimensional (2D) (Menapace, paragraph 58 teaches “fixed 2D noise tensor… is mapped to a plane of features”); this shows diffusion model conditioned on noise that is in 2D and it would also be conditioned with the image as explained in claim 1 above and since Zakharov abstract mentions “from a two-dimensional image”;
However, the combination of Zakharov, Menapace and Gupta fails to explicitly teach
the triplanar diffusion model is associated with a second probabilistic distribution of the object along multiple orthogonal planes in a second stage;
However, Brumby teaches the triplanar diffusion model is associated with a second probabilistic distribution of the object along multiple orthogonal planes in a second stage (Brumby, paragraph 185 teaches “second stage model 2302 may be a Bayesian probabilistic model where the expected distribution is learned from the observed sequence of labels from previous observations”); this shows second model such as the triplanar diffusion model of Gupta being second stage and therefore having second probabilistic distribution of object, also, this would be along multiple orthogonal planes since Gupta fig. 1 shows three planes arranged orthogonally. Brumby is considered to be analogous art because it is reasonably pertinent to the problem faced by the inventor of second stage and second probabilistic distribution. Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify the combination of Zakharov, Menapace and Gupta with the second stage techniques of Brumby to enable the mapping system to smooth out variability (Brumby, paragraph 185). This would lead to more stable and realistic result by the smoothening and avoiding extreme outlier.
However, the combination of Zakharov, Menapace, Gupta and Brumby fails to explicitly teach and the triplanar diffusion model is conditioned with an ortho-NORF representation of the image in a triplanar space and the triplanar noise.
However, Gudivada teaches and the triplanar diffusion model is conditioned with an ortho-NORF representation of the image in a triplanar space and the triplanar noise (Gudivada, abstract teaches “At each recursion a set of orthonormal bases functions are extracted for a specific scale. A diffusion step is embedded at each scale in the QR decomposition”); this shows orthonormal data which means the triplanar diffusion model from Gupta would be conditioned on such since the triplanar diffusion model inputs normalized object in Gupta, first paragraph of page 4 “We normalize each mesh” and normalized reference frame (NORF) would be the orthonormal data referred to herein when viewed in combination because both are normalized, also, since Gupta fig. 1 shows a triplanar space this would be done in such alongside the aforementioned triplanar noise in claim 1. Gudivada is considered to be analogous art because it is reasonably pertinent to the problem faced by the inventor of ortho-normal data. Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify the combination of Zakharov, Menapace, Gupta and Brumby with the ortho-normal techniques of Gudivada to improve the recognition results (Gudivada, page 4, second paragraph). This would lead to a more accurately rendered and reconstructed full object as well.
Regarding claim 20, the method claim 20 recites similar limitations as system claim 9, and thus is rejected under similar rationale.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. Wnuk et al. (U.S. Patent Application Publication No. 2022/0292804), abstract teaches “canonical shape object also includes one or more reference PoVs indicating perspectives from which to analyze objects having the corresponding shape” and paragraph 64 teaches “reference key frame PoV 224A, which can be represented by a normal vector to a surface of target shape object”); this shows a reference frame for a object which is represented by a normal vector.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to NAUMAN U AHMAD whose telephone number is (703)756-5306. The examiner can normally be reached Monday - Friday 9:00am - 5:00pm.
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, Kee Tung can be reached at (571) 272-7794. 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.
/N.U.A./Examiner, Art Unit 2611
/KEE M TUNG/Supervisory Patent Examiner, Art Unit 2611
Prosecution Insights