CTNF 18/959,422 CTNF 100765 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. Status of the Claims Claims 1-20 are currently pending in the present application, with claims 1, 11, and 18 being independent. Information Disclosure Statement The information disclosure statement (IDS) submitted on 11/25/2024 have been considered by the examiner. 07-30-03-h AIA Claim Interpretation 07-30-03 AIA The following is a quotation of 35 U.S.C. 112(f): (f) Element in Claim for a Combination. – An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof. The following is a quotation of pre-AIA 35 U.S.C. 112, sixth paragraph: An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof. 07-30-05 The claims in this application are given their broadest reasonable interpretation using the plain meaning of the claim language in light of the specification as it would be understood by one of ordinary skill in the art. The broadest reasonable interpretation of a claim element (also commonly referred to as a claim limitation) is limited by the description in the specification when 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is invoked. As explained in MPEP § 2181, subsection I, claim limitations that meet the following three-prong test will be interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph: (A) the claim limitation uses the term “means” or “step” or a term used as a substitute for “means” that is a generic placeholder (also called a nonce term or a non-structural term having no specific structural meaning) for performing the claimed function; (B) the term “means” or “step” or the generic placeholder is modified by functional language, typically, but not always linked by the transition word “for” (e.g., “means for”) or another linking word or phrase, such as “configured to” or “so that”; and (C) the term “means” or “step” or the generic placeholder is not modified by sufficient structure, material, or acts for performing the claimed function. Use of the word “means” (or “step”) in a claim with functional language creates a rebuttable presumption that the claim limitation is to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites sufficient structure, material, or acts to entirely perform the recited function. Absence of the word “means” (or “step”) in a claim creates a rebuttable presumption that the claim limitation is not to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is not interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites function without reciting sufficient structure, material or acts to entirely perform the recited function. Claim limitations in this application that use the word “means” (or “step”) are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action. Conversely, claim limitations in this application that do not use the word “means” (or “step”) are not being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action. Such claim limitation(s) is are: “a step for encoding…“ in claim 18. Claim Rejections - 35 USC § 112 07-30-02 AIA The following is a quotation of 35 U.S.C. 112(b): (b) CONCLUSION. —The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention. The following is a quotation of 35 U.S.C. 112 (pre-AIA), second paragraph: The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention. 07-34-01 Claim 9 is rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention. 07-34-03 AIA The term “ near real time ” in claim 9 is a relative term which renders the claim indefinite. The term “ near real time ” is not defined by the claim, the specification does not provide a standard for ascertaining the requisite degree, and one of ordinary skill in the art would not be reasonably apprised of the scope of the invention. The examiner respectfully requests the applicant to clarify the scope of the claimed invention. Claim 9 will be examined as best understood by the examiner . Claim Rejections - 35 USC § 103 07-06 AIA 15-10-15 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. 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, 4, 7, 10-11, 14, and 18 is/are rejected under 35 U.S.C. 103 as being unpatentable over Liang et al ., ENVIDR: Implicit Differentiable Renderer with Neural Environment Lighting, International Conference on Computer Vision, March 23, 2023, pp. 79-89, hereinafter referred to as “Liang” . Regarding claim 1 , Liang discloses accessing multiple input images of a specular object with a scene in multiple viewing directions (Pg. 80, Section 1; ENVIDR, a new rendering and modeling framework for high-quality reconstructing and rendering of 3D objects with challenging specular reflections. Fig. 3 spheres with varying views, environments, and materials. Pg. 83, Section 4.2; we train our neural renderer using synthesized images of a sphere with various materials and environment lighting…); encoding far-field reflections of the scene on the specular object to obtain a first set of feature representations based on the multiple input images (Pg. 82, Section 4; The environment light MLP encodes the distant light probes of a specific environment into neural features that interact with surface geometry features (i.e., feature fusion) through diffuse/specular MLPs. Pg. 82, Section 4.1; we represent the environment light probes as a coordinate-based MLP, conditioned on light directions…the output of our environment MLP is a high-dimensional neural feature vector instead of a valid HDR pixel…neural environment features…Pg. 83-84, Section 5.1; to estimate unknown environment light from training images, we randomly initialize an environment MLP Eg and optimize it alongside the geometry model Fg.); encoding near-field interreflections of the scene on the specular object to obtain a second set of feature representations of the specular object based on the multiple input images (Pg. 80, Section 1; shiny surfaces may have inter-reflections that cause apparent view-dependent indirect illumination…we approximate the incoming radiance from inter-reflections by marching rays along the surface-reflected view directions. We additionally propose a color blending model that converts the approximated incoming radiance into indirect illumination…Pg. 84, Section 5.2; focus on synthesizing inter-reflection on reflective surfaces with predicted roughness lower than a threshold ps…perform additional ray-marching (one-bounce) along the reflected view directions (see Fig. 6)…we introduce another color encoding MLP Eref to convert rendered reflected ray color into neural features frefenv compatible with our specular MLP Rs…Eq. (13)…); determining a set of specular color values of the specular object based on the first set of feature representations and the second set of feature representations using a decoding algorithm (Pg. 82, Section 4; we use three MLPs: environment light MLP E(·), diffuse MLP Rd(·), and specular MLP Rs(·) in place of NeRF's directional MLP…our diffuse MLP and specular MLP implicitly learn the rendering rules for the corresponding color components. Pg. 82-83, Section 4.1; given a light direction w^ and a roughness value p, our environment MLP E returns a neural feature vector fenv; Eq. (5)…environment features, geometry features, and the dot product between ωo and nˆ are then combined as the input to the specular MLP Rs: Eq. (7). Pg. 84, Section 5.2; Eq. (13)…Eq. (14). Examiner's note: ENVIDR's specular MLP Rs is the decoding algorithm, cs, cref, and/or the combined new specular color c's is the set of specular color values, The first feature set is fsenv from the environment MLP, the second feature set is frefenv from the inter-reflection color encoding MLP ); and providing a specular object representation at least based on the set of specular color values using a neural rendering algorithm (Pg. 83, Section 4.1; the synthesized diffuse and specular colors after volume rendering (Eq. 2) are additively combined in the linear space and then converted to sRGB space with gamma tone mapping: Eq. (8). Pg. 84, Section 5.2; To blend the rendered indirect illumination into our final rendering results, we let the geometry MLP Fg to additionally predict a blending factor η ∈ [0,1] (with Sigmoid activation) to combine the original direct specular color cs and indirect specular color cref into new specular color c's: Eq. (14)). Liang does not expressly disclose a method performed by one or more processing devices. Official Notice is taken, however, that it is well known and conventional to implement machine-learning model operations, including neural radiance fields (NeRFs), encoding and decoding operations, and volume rendering techniques, as computer-executable instructions stored on a non-transitory computer-readable storage medium and executed by one or more processors coupled to the memory. Such memory conventionally stores model parameters/executable instructions, while the processor conventionally executes the 3D scene modeling and reconstruction of neural feature outputs. Such implementation would have been an implied or otherwise obvious computer implementation of the disclosed model operations according to standard programming and machine-learning techniques, yielding predictable result of causing a computer processor to perform the recited neural-network operations (KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398,417 (2007)). Regarding claim 4 , Liang discloses the method of claim 1, and further discloses wherein determining the set of specular color values of the specular object based on the first set of feature representations and the second set of feature representations using a decoding algorithm comprising (Pg. 82, Section 4; we use three MLPs: environment light MLP E(·), diffuse MLP Rd(·), and specular MLP Rs(·) in place of NeRF's directional MLP…our diffuse MLP and specular MLP implicitly learn the rendering rules for the corresponding color components. Pg. 82-83, Section 4.1; given a light direction w^ and a roughness value p, our environment MLP E returns a neural feature vector fenv; Eq. (5)…environment features, geometry features, and the dot product between ωo and nˆ are then combined as the input to the specular MLP Rs: Eq. (7). Pg. 84, Section 5.2; Eq. (13)…Eq. (14). Examiner's note: ENVIDR's specular MLP Rs is the decoding algorithm, cs, cref, and/or the combined new specular color c's is the set of specular color values, The first feature set is fsenv from the environment MLP, the second feature set is frefenv from the inter-reflection color encoding MLP ) : decoding the first set of feature representations to a first set of specular color values representing far-field reflections using the decoding algorithm (Pg. 82-83, Section 4.1; query environment MLP E for the environment feature vector fsenv = E(wr, p). Environment features, geometry features, and the dot product between wo and n^ are then combined as the input to the specular MLP Rs: Eq. (7)…); decoding the second set of feature representations to a second set of specular color values representing near-field reflections using the decoding algorithm (Pg. 84, Section 5.2; color encoding MLP Eref to convert rendered reflected ray color into neural feature frefenv compatible with our specular MLP Rs. The rendered indirect illumination cref is then output by: Eq. (13)); and blending the first set of specular color values and the second set of specular color values to obtain an aggregate set of specular color values of the specular object (Pg. 84, Section 5.2; geometry MLP Fg to addionally predict a blending factor η ∈ [0,1] (with Sigmoid activation) to combine the original direct specular color cs and indirect specular color cref into new specular color c's: Edq. (14)). Regarding claim 7, Liang discloses the method of claim 1, and further discloses wherein the specular object representation further comprises an SDF-based geometry model (Pg. 80, Section 1; ENVIDR…2) an SDF-based neural surface model that represents the scene and interacts with the neural renderer…employs an SDF-based neural representation. Pg. 83, Section 5.1; utilizes a hybrid neural SDF representation Fg with multi-resolution feature grids and hash encoding…fed into a shallow MLP to predict all of SDF s, roughness p, and geometry feature fgeo…). Regarding claim 10, Liang discloses the method of claim 1, and further discloses wherein the neural rendering algorithm comprises a neural radiance fields (NeRF) algorithm (Fig. 4 and Pg. 80, Section 1; ENVIDR…a novel neural renderer…Pg. 83, Section 5.1; utilizes a hybrid neural SDF representation Fg with multi-resolution feature grids and has encoding for the efficient learning and rendering of scene surfaces). Regarding claim 11, Liang discloses accessing multiple input images of a specular object with a scene in multiple viewing directions (Pg. 80, Section 1; ENVIDR, a new rendering and modeling framework for high-quality reconstructing and rendering of 3D objects with challenging specular reflections. Fig. 3 spheres with varying views, environments, and materials. Pg. 83, Section 4.2; we train our neural renderer using synthesized images of a sphere with various materials and environment lighting…); encoding far-field reflections of the scene on the specular object to obtain a first set of feature representations based on the multiple input images (Pg. 82, Section 4; The environment light MLP encodes the distant light probes of a specific environment into neural features that interact with surface geometry features (i.e., feature fusion) through diffuse/specular MLPs. Pg. 82, Section 4.1; we represent the environment light probes as a coordinate-based MLP, conditioned on light directions…the output of our environment MLP is a high-dimensional neural feature vector instead of a valid HDR pixel…neural environment features…Pg. 83-84, Section 5.1; to estimate unknown environment light from training images, we randomly initialize an environment MLP Eg and optimize it alongside the geometry model Fg.); encoding near-field interreflections of the scene on the specular object to obtain a second set of feature representations of the specular object based on the multiple input images (Pg. 80, Section 1; shiny surfaces may have inter-reflections that cause apparent view-dependent indirect illumination…we approximate the incoming radiance from inter-reflections by marching rays along the surface-reflected view directions. We additionally propose a color blending model that converts the approximated incoming radiance into indirect illumination…Pg. 84, Section 5.2; focus on synthesizing inter-reflection on reflective surfaces with predicted roughness lower than a threshold ps…perform additional ray-marching (one-bounce) along the reflected view directions (see Fig. 6)…we introduce another color encoding MLP Eref to convert rendered reflected ray color into neural features frefenv compatible with our specular MLP Rs…Eq. (13)…); determining a set of specular color values of the specular object based on the first set of feature representations and the second set of feature representations using a decoding algorithm (Pg. 82, Section 4; we use three MLPs: environment light MLP E(·), diffuse MLP Rd(·), and specular MLP Rs(·) in place of NeRF's directional MLP…our diffuse MLP and specular MLP implicitly learn the rendering rules for the corresponding color components. Pg. 82-83, Section 4.1; given a light direction w^ and a roughness value p, our environment MLP E returns a neural feature vector fenv; Eq. (5)…environment features, geometry features, and the dot product between ωo and nˆ are then combined as the input to the specular MLP Rs: Eq. (7). Pg. 84, Section 5.2; Eq. (13) …Eq. (14). Examiner's note: ENVIDR's specular MLP Rs is the decoding algorithm, cs, cref, and/or the combined new specular color c's is the set of specular color values, The first feature set is fsenv from the environment MLP, the second feature set is frefenv from the inter-reflection color encoding MLP ); and providing a specular object representation at least based on the set of specular color values using a neural rendering algorithm (Pg. 83, Section 4.1; the synthesized diffuse and specular colors after volume rendering (Eq. 2) are additively combined in the linear space and then converted to sRGB space with gamma tone mapping: Eq. (8). Pg. 84, Section 5.2; To blend the rendered indirect illumination into our final rendering results, we let the geometry MLP Fg to additionally predict a blending factor η ∈ [0,1] (with Sigmoid activation) to combine the original direct specular color cs and indirect specular color cref into new specular color c's: Eq. (14)). Liang does not expressly disclose a system, comprising: a memory component, a processing device coupled to the memory component, the processing device to perform the recited operations. Official Notice is taken, however, that it is well known and conventional to implement machine-learning model operations, including neural radiance fields (NeRFs), encoding and decoding operations, and volume rendering techniques, as computer-executable instructions stored on a non-transitory computer-readable storage medium and executed by one or more processors coupled to the memory. Such memory conventionally stores model parameters/executable instructions, while the processor conventionally executes the 3D scene modeling and reconstruction of neural feature outputs. Such implementation would have been an implied or otherwise obvious computer implementation of the disclosed model operations according to standard programming and machine-learning techniques, yielding predictable result of causing a computer processor to perform the recited neural-network operations (KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398,417 (2007)). Regarding claim 14 , claim 14 has similar limitations as of claim 4, except it is a system claim (Official Notice), therefore it is rejected under the same rationale as claim 4. Regarding claim 18 , Liang discloses accessing multiple input images of a specular object with a scene in multiple viewing directions (Pg. 80, Section 1; ENVIDR, a new rendering and modeling framework for high-quality reconstructing and rendering of 3D objects with challenging specular reflections. Fig. 3 spheres with varying views, environments, and materials. Pg. 83, Section 4.2; we train our neural renderer using synthesized images of a sphere with various materials and environment lighting…); a step for encoding far-field reflections of the scene on the specular object to obtain a first set of feature representations based on the multiple input images (Pg. 82, Section 4; The environment light MLP encodes the distant light probes of a specific environment into neural features that interact with surface geometry features (i.e., feature fusion) through diffuse/specular MLPs. Pg. 82, Section 4.1; we represent the environment light probes as a coordinate-based MLP, conditioned on light directions…the output of our environment MLP is a high-dimensional neural feature vector instead of a valid HDR pixel…neural environment features…Pg. 83-84, Section 5.1; to estimate unknown environment light from training images, we randomly initialize an environment MLP Eg and optimize it alongside the geometry model Fg.); a step for encoding near-field interreflections of the scene on the specular object to obtain a second set of feature representations of the specular object based on the multiple input images (Pg. 80, Section 1; shiny surfaces may have inter-reflections that cause apparent view-dependent indirect illumination…we approximate the incoming radiance from inter-reflections by marching rays along the surface-reflected view directions. We additionally propose a color blending model that converts the approximated incoming radiance into indirect illumination…Pg. 84, Section 5.2; focus on synthesizing inter-reflection on reflective surfaces with predicted roughness lower than a threshold ps…perform additional ray-marching (one-bounce) along the reflected view directions (see Fig. 6)…we introduce another color encoding MLP Eref to convert rendered reflected ray color into neural features frefenv compatible with our specular MLP Rs…Eq. (13)…); providing a specular object representation at least based on a set of specular color values (Pg. 83, Section 4.1; the synthesized diffuse and specular colors after volume rendering (Eq. 2) are additively combined in the linear space and then converted to sRGB space with gamma tone mapping: Eq. (8). Pg. 84, Section 5.2; To blend the rendered indirect illumination into our final rendering results, we let the geometry MLP Fg to additionally predict a blending factor η ∈ [0,1] (with Sigmoid activation) to combine the original direct specular color cs and indirect specular color cref into new specular color c's: Eq. (14)) determined based on the first set of feature representations and the second set of feature representations (Pg. 82, Section 4; we use three MLPs: environment light MLP E(·), diffuse MLP Rd(·), and specular MLP Rs(·) in place of NeRF's directional MLP…our diffuse MLP and specular MLP implicitly learn the rendering rules for the corresponding color components. Pg. 82-83, Section 4.1; given a light direction w^ and a roughness value p, our environment MLP E returns a neural feature vector fenv; Eq. (5)…environment features, geometry features, and the dot product between ωo and nˆ are then combined as the input to the specular MLP Rs: Eq. (7). Pg. 84, Section 5.2; Eq. (13)…Eq. (14). Examiner's note: ENVIDR's specular MLP Rs is the decoding algorithm, cs, cref, and/or the combined new specular color c's is the set of specular color values, The first feature set is fsenv from the environment MLP, the second feature set is frefenv from the inter-reflection color encoding MLP ). Liang does not expressly disclose a non-transitory computer-readable medium, storing executable instructions, which when executed by a processing device, cause the processing device to perform the recited operations. Official Notice is taken, however, that it is well known and conventional to implement machine-learning model operations, including neural radiance fields (NeRFs), encoding and decoding operations, and volume rendering techniques, as computer-executable instructions stored on a non-transitory computer-readable storage medium and executed by one or more processors coupled to the memory. Such memory conventionally stores model parameters/executable instructions, while the processor conventionally executes the 3D scene modeling and reconstruction of neural feature outputs. Such implementation would have been an implied or otherwise obvious computer implementation of the disclosed model operations according to standard programming and machine-learning techniques, yielding predictable result of causing a computer processor to perform the recited neural-network operations (KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398,417 (2007)) . 07-21-aia AIA Claim (s) 2, 9, and 12 is/are rejected under 35 U.S.C. 103 as being unpatentable over Liang et al ., ENVIDR: Implicit Differentiable Renderer with Neural Environment Lighting, International Conference on Computer Vision, March 23, 2023, pp. 79-89, hereinafter referred to as “Liang”, in view of Karis, Real Shading in Unreal Engine 4, Course: Physically Based Shading in Theory and Practice, Volume 4, No. 3, 2013, 59 pages . Regarding claim 2, Liang discloses the method of claim 1, but does not disclose encoding the far-field reflections of the scene on the specular object into a cubemap to obtain the first set of feature representations, wherein the first set of feature representations of the specular object comprises cubemap-based far-field feature representations. Karis discloses encoding the far-field reflections of the scene on the specular object into a cubemap to obtain the first set of feature representations, wherein the first set of feature representations of the specular object comprises cubemap-based far-field feature representations (Pg. 5-6, Section Pre-Filtered Environment Map; We pre-calculate the first sum for different roughness values and store the results in the mip-map levels of a cubemap). It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate Karis’ cubemap-based prefiltered environment maps into Liang’s far-field reflections/distant environment lighting. Liang uses distant light probes and IDE over input direction and roughness to learn high-frequency environment feature vectors, and Karis provides a conventional graphics technique of representing the far-field reflection information as cubemap mip levels. Combining allows direction-dependent environment reflection information to be sampled efficiently for rendering outputs, and cubemaps are a conventional and efficient representation for specular image-based lighting. Regarding claim 9, Liang discloses the method of claim 1, but does not disclose providing the specular object representation in real time or near real time. Karis discloses providing the specular object representation in real time or near real time (Pg. 1, Section Introduction; Real-time Performance…Pg. 9, Section Material Model; real-time rendering. Pg. 12, Section Lighting Model; for efficiency…in real-time). It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to accelerate Liang’s neural rendering pipeline using a real-time specular object representation. Doing so enables interactive real-time novel-view rendering while preserving view-dependent effects including specularity. Regarding claim 12 , claim 12 has similar limitations as of claim 2, except it is a system claim (Official Notice), therefore it is rejected under the same rationale as claim 2 . 07-21-aia AIA Claim (s) 3 and 13 is/are rejected under 35 U.S.C. 103 as being unpatentable over Liang et al ., ENVIDR: Implicit Differentiable Renderer with Neural Environment Lighting, International Conference on Computer Vision, March 23, 2023, pp. 79-89, hereinafter referred to as “Liang”, in view of Crassin et al., Interactive Indirect Illumination Using Voxel Cone Tracing, Computer Graphics Forum, September 2011, pp. 1-10, hereinafter referred to as “Crassin” . Regarding claim 3, Liang discloses the method of claim 1, but does not disclose encoding the near-field interreflections of the scene by cone-tracing a spatial feature grid to obtain the second set of feature representations, wherein the second set of feature representations of the specular object comprises cone-traced near-field feature representations. Crassin discloses encoding the near-field interreflections of the scene by cone-tracing a spatial feature grid to obtain the second set of feature representations, wherein the second set of feature representations of the specular object comprises cone-traced near-field feature representations (Section 1; a pre-filtered hierarchical voxel representation of the scene geometry…pre-filtered representation to quickly estimate visibility and integrate incoming indirect energy splatted in the structure from the light sources, using a new approximate voxel cone tracing technique…Section 3; employ an approximate cone tracing to perform a final gathering, sending out a few cones over the hemisphere to collect illumination distributed in the octree…a few large cones (~5) estimate the diffuse energy coming from the scene, while a tight cone in the reflected direction with respect to the viewpoint captures the specular component…). It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to implement Liang’s reflected-ray interreflection estimation using Crassin’s voxel cone tracing through a spatial voxel feature grid. Doing so provides a fast approximation for visibility and incoming energy for indirect illumination (Crassin Section 1; …pre-filtered representation to quickly estimate visibility and integrate incoming indirect energy splatted in the structure from the light sources…), and using a known cone-tracing technique for estimate indirect illumination in a spatial representation yields predictable results in improved efficiency when estimating near-field interreflection radiance. Regarding claim 13 , claim 13 has similar limitations as of claim 3, except it is a system claim (Official Notice), therefore it is rejected under the same rationale as claim 3 . 07-21-aia AIA Claim (s) 5-6, 15-16, and 19 is/are rejected under 35 U.S.C. 103 as being unpatentable over Liang et al ., ENVIDR: Implicit Differentiable Renderer with Neural Environment Lighting, International Conference on Computer Vision, March 23, 2023, pp. 79-89, hereinafter referred to as “Liang”, in view of Muller et al., Instant Neural Graphics Primitives with a Multiresolution Hash Encoding, Association for Computing Machinery Transactions on Graphics, Volume 41, No. 4, July 2022, pp. 1-15, hereinafter referred to as “Muller” . Regarding claim 5, Liang discloses the method of claim 1, and further discloses wherein determining the set of specular color values of the specular object based on the first set of feature representations and the second set of feature representations using a decoding algorithm comprising : decoding the aggregate set of feature representations to obtain an aggregated set of specular color values of the specular object (Pg. 82, Section 4; environment light MLP encodes the distant light probes of a specific environment into neural features that interact with surface geometry features (i.e., Feature fusion) through diffuse/specular MLPs. Pg. 82-83, Section 4.1; Environment features, geometry features, and the dot product between wo and n^ are then combined as the input to the specular MLP Rs: Eq. (7). Pg. 84, Section 5.2; geometry MLP Fg to addionally predict a blending factor η ∈ [0,1] (with Sigmoid activation) to combine the original direct specular color cs and indirect specular color cref into new specular color c's: Eq. (14)). Liang does not appear to explicitly disclose blending the first set of feature representations and the second set of feature representations to obtain an aggregate set of feature representations. Muller discloses blending the first set of feature representations and the second set of feature representations to obtain an aggregate set of feature representations (Pg. 102:3, Section 2; we use multiple separate hash tables indexed at different resolutions, whose interpolated outputs are concatenated before being passed through the MLP. Fig. 3; multiresolution hash encoding….look up the corresponding F-dimensional feature vectors from the hash tables 𝜃 t…linearly interpolate them…concatenate the result of each level, as well as auxiliary inputs s 𝜉∈ R 𝐸 , producing the encoded MLP input 𝑦∈ R 𝐿𝐹 +E, which is evaluated…); It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to combine Liang’s first neural environment features and second interreflection neural features with Muller’s concatenation of interpolated feature vectors before inputting them to the specular MLP decoder to obtain aggregate specular color values. Liang’s model already uses neural feature fusion as inputs to its rendering MLPs, and Muller teaches that interpolated feature vectors and auxiliary inputs are concatenated to form an encoded MLP input evaluated by a neural network, therefore providing a known neural graphics technique for combining multiple feature representations before decoding would’ve yielded predictable results in improved decoding accuracy, robustness, and efficiency for real-time applications such as gaming or AR/VR. Regarding claim 6, Liang discloses the method of claim 1, but does not disclose wherein the decoding algorithm comprises a multi-layer perceptron (MLP) network, wherein the MLP network comprises two layers with a width of 64 Muller discloses wherein the decoding algorithm comprises a multi-layer perceptron (MLP) network, wherein the MLP network comprises two layers with a width of 64 (Pg. 102:6, Section 4; In all tasks…we use an MLP with two hidden layers that have a width of 64 neurons…Fig. 10; 𝑁 layers = 2 and 𝑁 neurons = 64 and Section 5.4 Model Architecture; NeRF model consists of two concatenated MLPs: a density MLP followed by a color MLP…The color MLP adds view-dependent color variation. Its input is the concatenation of…its output is an RGB color triplet…our results were generated with a…2-hidden-layer color MLP, both 64 neurons wide…). It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to incorporate Muller’s two-hidden-layer, width-64 MLP architecture for Liang’s specular MLP decoder. Liang uses MLPs to decode neural features into specular color, and Muller teaches that small MLPs combined with multiresolution feature encodings provide efficient neural graphics primitives while maintaining quality (Muller Abstract: permits the use of a smaller network without sacrificing quality, thus significantly reducing the number of floating point and memory access operations…focus on minimizing wasted bandwidth and compute operations. We achieve a combined speedup of several orders of magnitude, enabling training of high-quality neural graphics primitives in a matter of seconds, and rendering in tens of milliseconds at a resolution of 1920×1080). Regarding claim 15 , claim 15 has similar limitations as of claim 5, except it is a system claim (Official Notice), therefore it is rejected under the same rationale as claim 5. Regarding claim 16 , claim 16 has similar limitations as of claim 6, except it is a system claim (Official Notice), therefore it is rejected under the same rationale as claim 6. Regarding claim 19 , claim 19 has similar limitations as of claim 5, except it is a CRM claim (Official Notice), therefore it is rejected under the same rationale as claim 5 . 07-21-aia AIA Claim (s) 8, 17, and 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Liang et al ., ENVIDR: Implicit Differentiable Renderer with Neural Environment Lighting, International Conference on Computer Vision, March 23, 2023, pp. 79-89, hereinafter referred to as “Liang”, in view of Kingma et al., Adam: A Method for Stochastic Optimization, International Conference on Learning Representations, Conference paper at ICLR 2015, Available Online at: https://arxiv.org/pdf/1412.6980.pdf, January 30, 2017, pp. 1-15, hereinafter referred to as “Kingma” . Regarding claim 8, Liang discloses the method of claim 7, and further discloses the SDF-based geometry model of the specular object model based on the set of specular color values (Pg. 80, Section 1; ENVIDR…2) an SDF-based neural surface model that represents the scene and interacts with the neural renderer…employs an SDF-based neural representation. Pg. 83, Section 5.1; utilizes a hybrid neural SDF representation Fg with multi-resolution feature grids and hash encoding…fed into a shallow MLP to predict all of SDF s, roughness p, and geometry feature fgeo…). Liang does not disclose optimizing the SDF-based geometry model of the specular object model based on the set of specular color values using an Adam optimizer. Kingma discloses optimizing the SDF-based geometry model of the specular object model based on the set of specular color values using an Adam optimizer (Section 1; Adam, a method for efficient stochastic optimization that only requires first-order gradients with little memory requirement. The method computes individual adaptive learning rates for different parameters from estimates of first and second moments of the gradients…Algorithm 1…Fig. 2). It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to use Kingma’s Adam method to optimize Liang’s neural SDF/MLP parameters. Adam is a well-known stochastic optimization technique for training neural networks, therefore applying a known neural network optimizer to SDF/MLP training would have been a predictable implementation choice (Kingma Fig. 2 and Section 1; Some of Adam’s advantages are that the magnitudes of parameter updates are invariant to rescaling of the gradient, its stepsizes are approximately bounded by the stepsize hyperparameter, it does not require a stationary objective, it works with sparse gradients, and it naturally performs a form of step size annealing…. Adam is a versatile algorithm that scales to large-scale high-dimensional machine learning problems). Regarding claim 17, Liang discloses the system of claim 11, and further discloses wherein the specular object representation further comprises an SDF-based geometry model (Pg. 80, Section 1; ENVIDR…2) an SDF-based neural surface model that represents the scene and interacts with the neural renderer…employs an SDF-based neural representation. Pg. 83, Section 5.1; utilizes a hybrid neural SDF representation Fg with multi-resolution feature grids and hash encoding…fed into a shallow MLP to predict all of SDF s, roughness p, and geometry feature fgeo…). Liang does not disclose optimizing the SDF-based geometry model of the specular object model based on the set of specular color values using an Adam optimizer. Kingma discloses optimizing the SDF-based geometry model of the specular object model based on the set of specular color values using an Adam optimizer (Section 1; Adam, a method for efficient stochastic optimization that only requires first-order gradients with little memory requirement. The method computes individual adaptive learning rates for different parameters from estimates of first and second moments of the gradients…Algorithm 1…Fig. 2). Liang and Kingma are combined for the reasons set forth above with respect to claim 8. Regarding claim 20 , claim 20 has similar limitations as of claim 17, except it is a CRM claim (Official Notice), therefore it is rejected under the same rationale as claim 17. Conclusion Any inquiry concerning this communication or earlier communications from the examiner should be directed to JENNY NGAN TRAN whose telephone number is (571)272-6888. The examiner can normally be reached Mon-Thurs 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, Alicia Harrington can be reached at (571) 272-2330. 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. /JENNY N TRAN/Examiner, Art Unit 2615 /ALICIA M HARRINGTON/Supervisory Patent Examiner, Art Unit 2615 Application/Control Number: 18/959,422 Page 2 Art Unit: 2615 Application/Control Number: 18/959,422 Page 3 Art Unit: 2615 Application/Control Number: 18/959,422 Page 4 Art Unit: 2615 Application/Control Number: 18/959,422 Page 5 Art Unit: 2615 Application/Control Number: 18/959,422 Page 6 Art Unit: 2615 Application/Control Number: 18/959,422 Page 7 Art Unit: 2615 Application/Control Number: 18/959,422 Page 8 Art Unit: 2615 Application/Control Number: 18/959,422 Page 9 Art Unit: 2615 Application/Control Number: 18/959,422 Page 10 Art Unit: 2615 Application/Control Number: 18/959,422 Page 11 Art Unit: 2615 Application/Control Number: 18/959,422 Page 12 Art Unit: 2615 Application/Control Number: 18/959,422 Page 13 Art Unit: 2615 Application/Control Number: 18/959,422 Page 14 Art Unit: 2615 Application/Control Number: 18/959,422 Page 15 Art Unit: 2615 Application/Control Number: 18/959,422 Page 16 Art Unit: 2615 Application/Control Number: 18/959,422 Page 17 Art Unit: 2615 Application/Control Number: 18/959,422 Page 18 Art Unit: 2615 Application/Control Number: 18/959,422 Page 19 Art Unit: 2615 Application/Control Number: 18/959,422 Page 20 Art Unit: 2615 Application/Control Number: 18/959,422 Page 21 Art Unit: 2615 Application/Control Number: 18/959,422 Page 22 Art Unit: 2615
Prosecution Insights