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 .
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 11/10/2023 and 06/09/2023 have been entered and considered. Initialed copies of the PTO-1449 by the Examiner are attached.
Election/Restrictions
Applicant's election with traverse of Group II consisting of claims 9-16, 18 and 20 in the reply filed on March 2, 2026 is acknowledged. There were no arguments presented for the traversal. The requirement is still deemed proper and is therefore made FINAL.
Claim Rejections - 35 USC § 102
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 the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claims 1-2 and 7-8 are rejected under 35 U.S.C. 102(a)(2) as being anticipated by anticipated Ost et al (NPL titled: Neural Point Light Fields) [cited in the IDS]
Regarding independent claim 1 Ost teaches a method for training a neural network model (trained end-to-end with the full light field rendering – see section 3.1, [p][003]), comprising: acquiring an image captured by a camera about a scene (a camera-lidar sensor setup typical in robotic and automotive contexts [10], at time step i, the proposed method learns an RGB frame Ii as input and the corresponding point cloud capture Pi – see section 3, [p][001]); determining a plurality of rays at least according to parameters of the camera ([g]iven a set of points Pi = fx0; :::; xNgiwith xk 2 R3, their encoded features lk 2 R6128, and a camera view Ci, defined by its intrinsic K, extrinsic Ei – see section 3.2, [p][001] and [f]or each predicted ray color ˆ C(rj) we can compute the mean-squared error image loss – see section 3.3, [p][003]); determining a plurality of sampling points according to a relative positional relationship between the rays and a point cloud (camera-lidar sensor setup typical in robotic and automotive contexts [10], at time step i, the proposed method learns an RGB frame Ii as input and the corresponding point cloud capture Pi. To learn a light field embedded on the point clouds corresponding to a video sequence – see section 3, [p][001]), wherein the point cloud is associated with a part of the scene (see section 3, [p][001] and Fig 2 depicts K closest points); determining color information of pixels of the image which correspond to the sampling points (multi-head attention learns to predict a weight for all Vk,j given Kk,j, for each selected point ray pair (k,j) and query ray Q – see section 3.2, [p][004] and color of the ray of the scene and thus color of the query pixel - section 3.3); and training the neural network model with the sampling points and the color information of the pixels ([t]raining All model parameters, namely θResNet18, θK, θV , θQ, θattn and θLF, are jointly optimized by minimizing the loss in Eq. 10 using the Adam optimizer [16] with a linear learning rate decay, where at each step we randomly sample 8192 rays from a small batch of frames – see section 3.3 [p][002]).
Regarding claim 2, Ost teaches the method according to claim 1, further comprising: determining content of the image which is associated with the part of the scene (proposes to separate background and foreground scene components – see section 2 [p][002] and dynamic object support – see section 1, [p][005]), wherein determining a plurality of rays at least according to parameters of the camera when capturing the image (g]iven a set of points Pi = fx0; :::; xNgiwith xk 2 R3, their encoded features lk 2 R6128, and a camera view Ci, defined by its intrinsic K, extrinsic Ei – see section 3.2, [p][001]) comprises: determining the plurality of rays according to the parameters of the camera when capturing the image and the content of the image which is associated with the part of the scene (g]iven a set of points Pi = fx0; :::; xNgiwith xk 2 R3, their encoded features lk 2 R6128, and a camera view Ci, defined by its intrinsic K, extrinsic Ei – see section 3.2, [p][001].
Regarding claim 7, Ost teaches the method according to claim 1, further comprising: generating a representation of the part of the scene according to the point cloud, wherein determining a plurality of sampling points according to a relative positional relationship between the rays and a point cloud (guide the training of implicit scene representations [6] or offer a scaffold for learned features – see section 2, [p][002] ) comprises: determining intersection points of the rays with the representation as the sampling points (optimizing point locations from an ini tial point cloud, together with their novel view synthesis pipeline - see section 2, [p][002]).
Regarding claim 8, Ost teaches The method according to claim 7, wherein the point cloud is an aggregated point cloud (aggregate the features that are relevant for reconstructing the local light field around each ray – see section 3.2, [p][001]), the method further comprising: acquiring a sequence of point clouds associated with the part of the scene (Given a set of points Pi = {x0,...,xN}iwith xk ∈ R3, their encoded features lk ∈ R6× 128, and a camera view Ci, defined by its intrinsic – see section 3.2, [p][001]); registering the point clouds of the sequence (see section 3.2, [p][001])); and superimposing the registered point clouds with each other to obtain the aggregated point cloud (we cast a set of rays Ri into the scene using a pinhole camera model - see section 3.2, [p][001]).
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 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 of this title, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 3, 17 and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Ost et al (NPL titled: Neural Point Light Fields) [cited in the IDS] in view of Mase (Pub No.: US20130208984A1).
Regarding claim 3, Ost teaches the method according to claim 2, Ost teach wherein the part of the scene is a first part of the scene, wherein determining content of the image which is associated with the part of the scene comprises (proposes to separate background and foreground scene components – see section 2 [p][002] and dynamic object support – see section 1, [p][005]):
Ost does not explicitly teach determining content of the image which is associated with a second part of the scene, the second part being different from the first part; and removing the content of the image which is associated with the second part of the scene from the image.
However, Mase explicitly teaches determining content of the image which is associated with a second part of the scene ([t]he second scene determination means 6 also has a function of comparing the generated second content related data with one or a plurality of pieces of second reference content related data to determine a secondary object included in the input content – see [p][0030]), the second part being different from the first part (see [p][0029-0030]); and removing the content of the image which is associated with the second part of the scene from the image (generates second content related data by eliminating the influence of the area which is determined that the primary object is present from the first content related data extracted from the input content – see [p][0036]).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention was made to combine the teachings of Ost of a method for training a neural network model with the teachings of Mase for determining content of the image which is associated with a second part of the scene, the second part being different from the first part; and removing the content of the image which is associated with the second part of the scene from the image.
Wherein having Ost determining content of the image which is associated with a second part of the scene, the second part being different from the first part; and removing the content of the image which is associated with the second part of the scene from the image.
The motivation behind the modification would have been for synthesize photo-realistic images for novel views of small scenes while reduce the number of pieces of reference data required for determining a scene since both Ost and Mase image processing wherein Ost synthesize photo-realistic images for novel views of small scenes while Mase reduces the number of pieces of reference data required for determining a scene (Please see Ost et al (NPL titled: Neural Point Light Fields), abstract and Mase (Pub No.: US20130208984A1), [p][0014]).
Regarding claim 17, Ost does not explicitly teach an electronic device, comprising: a processor; and a memory storing instructions that, when executed by the processor, cause the processor to perform the method according to claim 1.
However, Mase explicitly teaches an electronic device, comprising: a processor (see [p][0040]); and a memory storing instructions that, when executed by the processor, cause the processor to perform the method according to claim 1 (see [p][0040]).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention was made to combine the teachings of Ost of a method for training a neural network model with the teachings of Mase for an electronic device, comprising: a processor; and a memory storing instructions that, when executed by the processor, cause the processor to perform the method according to claim 1.
Wherein having Ost an electronic device, comprising: a processor; and a memory storing instructions that, when executed by the processor, cause the processor to perform the method according to claim 1.
The motivation behind the modification would have been for synthesize photo-realistic images for novel views of small scenes while reduce the number of pieces of reference data required for determining a scene since both Ost and Mase image processing wherein Ost synthesize photo-realistic images for novel views of small scenes while Mase reduces the number of pieces of reference data required for determining a scene (Please see Ost et al (NPL titled: Neural Point Light Fields), abstract and Mase (Pub No.: US20130208984A1), [p][0014]).
Regarding claim 19, Ost does not explicitly teach a non-transitory computer-readable storage medium storing instructions that, when executed by a processor of a computing device, cause the computing device to perform the method according to claim 1.
However, Mase explicitly teaches a non-transitory computer-readable storage medium storing instructions that, when executed by a processor of a computing device, cause the computing device to perform the method according to claim 1 (see [p][0040]).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention was made to combine the teachings of Ost of a method for training a neural network model with the teachings of Mase for a non-transitory computer-readable storage medium storing instructions that, when executed by a processor of a computing device, cause the computing device to perform the method according to claim 1.
Wherein having Ost a non-transitory computer-readable storage medium storing instructions that, when executed by a processor of a computing device, cause the computing device to perform the method according to claim 1.
The motivation behind the modification would have been for synthesize photo-realistic images for novel views of small scenes while reduce the number of pieces of reference data required for determining a scene since both Ost and Mase image processing wherein Ost synthesize photo-realistic images for novel views of small scenes while Mase reduces the number of pieces of reference data required for determining a scene (Please see Ost et al (NPL titled: Neural Point Light Fields), abstract and Mase (Pub No.: US20130208984A1), [p][0014]).
Claim 4 is rejected under 35 U.S.C. 103 as being unpatentable over Ost et al (NPL titled: Neural Point Light Fields) [cited in the IDS] in view of
Remates et al (NPL titled: Urban Radiance Fields) [cited in IDS].
Regarding claim 4, Ost teaches the method according to claim 2, wherein the part is a static part of the scene which comprises one or more static objects of the scene (static scenes – see Table 2 description), wherein determining content of the image which is associated with the part of the scene comprises: determining content of the image which is associated with a dynamic object of the scene (dynamic object support – see section 1, [p][005]).
Ost does not explicitly teach determining a projection of the dynamic object according to a moment when the image is captured, and removing the content associated with the dynamic object and content associated with the projection from the image.
Rematas explicitly teach determining a projection of the dynamic object according to a moment when the image is captured (images are posed automatically within a GPS system using structure from motion and GSP information allowing us to assemble camera ray with origin o and direction d corresponding to each point – see section 3, [p][002]), and removing the content associated with the dynamic object and content associated with the projection from the image (object that have motion such as people are masked out – see section 3, [p][003]).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention was made to combine the teachings of Ost of a method for training a neural network model with the teachings of Rematas for determining a projection of the dynamic object according to a moment when the image is captured, and removing the content associated with the dynamic object and content associated with the projection from the image.
Wherein having Ost determining a projection of the dynamic object according to a moment when the image is captured, and removing the content associated with the dynamic object and content associated with the projection from the image.
The motivation behind the modification would have been for synthesize photo-realistic images for novel views of small scenes while removing a moving object since both Ost and Rematas image processing wherein Ost synthesize photo-realistic images for novel views of small scenes while Rematas removed moving object by masking (Please see Ost et al (NPL titled: Neural Point Light Fields), abstract and Remates et al (NPL titled: Urban Radiance Fields), see section 3, [p][003]).
Claims 5-6 are rejected under 35 U.S.C. 103 as being unpatentable over Ost et al (NPL titled: Neural Point Light Fields) [cited in the IDS] in view of Yoon (Pub No.: US20180173239A1)
Regarding claim 5, Ost does not explicitly teach the method according to claim 1, further comprising: generating a grid comprising a plurality of grid points, mapping each point of the point cloud to a respective one of the plurality of grid points to obtain a plurality of point-cloud-mapped points, wherein determining a plurality of sampling points according to a relative positional relationship between the rays and a point cloud comprises: selecting a plurality of points on each of the rays, for each of the plurality of points on the ray: mapping the point to one of the plurality of grid points to obtain a ray-mapped point, determining whether the ray-mapped point is coincident with one of the plurality of point-cloud-mapped points, and in response to the ray-mapped point being coincident with the one of the plurality of point-cloud-mapped points, generating one of the plurality of sampling points according to one of the point on the ray, the point-cloud-mapped point, and a point of the point cloud which corresponds to the point-cloud-mapped point.
However, Yoon explicitly teaches generating a grid () comprising a plurality of grid points, mapping each point of the point cloud to a respective one of the plurality of grid points to obtain a plurality of point-cloud-mapped points (the mapping line generation unit 220 may generate a mapping line based on the point clouds obtained from the sensor 210 – see [p0067]), wherein determining a plurality of sampling points according to a relative positional relationship between the rays and a point cloud (At step 321, the super ray generation unit 230 may determine any one point cloud for each segment section with respect to point clouds located in each of a plurality of segment sections that divides the mapping line – see [p][0072]) comprises: selecting a plurality of points on each of the rays, for each of the plurality of points on the ray (see [p][0079]): mapping the point to one of the plurality of grid points to obtain a ray-mapped point, determining whether the ray-mapped point is coincident with one of the plurality of point-cloud-mapped points (, in order to identify point clouds mapped to the same super ray, a line segment that overlaps between the cell C and a slice (i.e., a slice 3 423) including the cell C may be generated as a mapping line – see [p][0080]), and in response to the ray-mapped point being coincident with the one of the plurality of point-cloud-mapped points, generating one of the plurality of sampling points according to one of the point on the ray, the point-cloud-mapped point, and a point of the point cloud which corresponds to the point-cloud-mapped point (determine a plurality of intersection points 405 and 406 connected from the sensor origin 401 in a slice 1 411, including the sensor origin 401, to both end points 403 and 404 of the cell box 402. Referring to an occupancy map 420, the mapping line generation unit 220 may determine a grid point (g1) 426 located between the determined intersection points 424 and 425 in the slice 1 411 including the sensor origin 401. Furthermore, the mapping line generation unit 220 may divide an initial mapping line into a plurality of segment sections by dividing the seed frustum into a plurality of sub-frustums based on the determined grid point 426. For example, if the sensor origin 401 and both end points 403 and 404 of the cell box 402 in the occupancy map 410 are to be connected, the intersection points 403 and 404 with the two line segments 407 and 408 and the cell box 402 in a slice 3 including the cell C may be generated as an initial mapping line. Furthermore, the initial mapping line may be divided into a plurality of segment sections by dividing the seed frustum. The initial mapping line starts from a single line segment representing a super ray, and the seed frustum is divided into a plurality of sub-frustums and thus the initial mapping line is divided into several segment sections. Accordingly, a line segment corresponding to each of the segment sections may correspond to each of a plurality of super rays – see [p][0081])
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention was made to combine the teachings of Ost of a method for training a neural network model with the teachings of Yoon for generating a grid comprising a plurality of grid points, mapping each point of the point cloud to a respective one of the plurality of grid points to obtain a plurality of point-cloud-mapped points, wherein determining a plurality of sampling points according to a relative positional relationship between the rays and a point cloud comprises: selecting a plurality of points on each of the rays, for each of the plurality of points on the ray: mapping the point to one of the plurality of grid points to obtain a ray-mapped point, determining whether the ray-mapped point is coincident with one of the plurality of point-cloud-mapped points, and in response to the ray-mapped point being coincident with the one of the plurality of point-cloud-mapped points, generating one of the plurality of sampling points according to one of the point on the ray, the point-cloud-mapped point, and a point of the point cloud which corresponds to the point-cloud-mapped point.
Wherein having Ost generating a grid comprising a plurality of grid points, mapping each point of the point cloud to a respective one of the plurality of grid points to obtain a plurality of point-cloud-mapped points, wherein determining a plurality of sampling points according to a relative positional relationship between the rays and a point cloud comprises: selecting a plurality of points on each of the rays, for each of the plurality of points on the ray: mapping the point to one of the plurality of grid points to obtain a ray-mapped point, determining whether the ray-mapped point is coincident with one of the plurality of point-cloud-mapped points, and in response to the ray-mapped point being coincident with the one of the plurality of point-cloud-mapped points, generating one of the plurality of sampling points according to one of the point on the ray, the point-cloud-mapped point, and a point of the point cloud which corresponds to the point-cloud-mapped point.
The motivation behind the modification would have been for synthesize photo-realistic images for novel views of small scenes while updating an occupancy map since both Ost and Yoon image processing wherein Ost synthesize photo-realistic images for novel views of small scenes while Yoon updates an occupancy map (Please see Ost et al (NPL titled: Neural Point Light Fields), abstract and Yoon (Pub No.: Pub No.: US20180173239A1), [p][0014]).
Regarding claim 6, Ost in view of Yoon explicitly teach the method according to claim 5, further comprising: Ost teaches storing the point-cloud-mapped point in a Hash table (see section 2, [p][002]).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Xu et al (Pub No.: 20240013477) discloses a scene modeling system receives a plurality of input two-dimensional (2D) images corresponding to a plurality of views of an object and a request to display a three-dimensional (3D) scene that includes the object. The scene modeling system generates an output 2D image for a view of the 3D scene by applying a scene representation model to the input 2D images. The scene representation model includes a point cloud generation model configured to generate, based on the input 2D images, a neural point cloud representing the 3D scene. The scene representation model includes a neural point volume rendering model configured to determine, for each pixel of the output image and using the neural point cloud and a volume rendering process, a color value. The scene modeling system transmits, responsive to the request, the output 2D image. Each pixel of the output image includes the respective determined color value.
Yan et al (Pub No.: 20230401837 ) discloses a method for training a neural network model and a method for generating an image. The method for training a neural network model includes: acquiring an image about a scene captured by a camera; determining a plurality of rays at least according to parameters of the camera when capturing the image; determining a plurality of sampling points according to a relative positional relationship between the rays and a point cloud, where the point cloud is associated with a part of the scene; determining color information of pixels of the image which correspond to the sampling points; and training the neural network model according to positions of the sampling points and the color information of the pixels.
Inquiries
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ANDRAE S ALLISON whose telephone number is (571)270-1052. The examiner can normally be reached on Monday-Friday 9am-5pm 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, Chineyere Wills-Burns, can be reached on (571) 272-9752. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/ANDRAE S ALLISON/Primary Examiner, Art Unit 2673
March 18, 2026