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 .
Status of the Claims
Claims 1-32 are currently pending in the present application, with claims 1, 28, and 31 being independent.
Claim Objections
Claim 30 objected to because of the following informalities: The recitation of “The method of claim 29,” should be amended to recite “The system of claim 29”. Appropriate correction is required.
Claim Rejections - 35 USC § 112
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.
Claims 22-24, and 27 is/are 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
Regarding claim 22, the term “new corrective term” in claim 22 is a relative term which renders the claim indefinite. The term “new” 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 claim does not specify what preexisting formulation lacks the visibility-gradient contribution or relative to what earlier term the correct term is “new” to. Further, the claim does not identify the mathematical quantity to which the corrective term is added, such as a rendering integral, derivative estimator, gradient, or loss function
Regarding claim 23, claim 23 recites the limitation "the gradient" in line 1. There is insufficient antecedent basis for this limitation in the claim. It is unclear whether “the gradient” could be “a missing visibility gradient,” a “continuous visibility gradient,” or some other gradient.
Regarding claim 24, the recitation of “an area of known support for integrating” does not identify what is being integrated or the integral process for which support is provided, or whether the claimed “Area” is a surface area, an area in directionally space, a solid-angle region, or a different area. Although the central ray and angle may define an angular region, the claim does not identify what is integrated over that region. Accordingly, the metes and bounds of the limitation cannot be determined with reasonable certainty.
Regarding claim 27, the recitation of “used to determine which neighbors can be used at minimal cost…” is unclear. The claim does not identify what the recited “neighbors” are neighbors of. For example, they could be neighboring rays, pixels, samples, cones, or geometric primitives. The claim also does not identify what information associated with those neighbors is” reused.” Therefore, one or ordinary skill in the art cannot determine the scope of the required neighbor and reuse operation.
The examiner respectfully requests the applicant to clarify the scope of the claimed invention.
Claims 22-24, and 27 will be examined as best understood by the examiner.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim(s) 1, 3-5, 7, 9-11, 14-15, 17, 19-22, 24-26, and 28-32 is/are rejected under 35 U.S.C. 103 as being unpatentable over Sawhney et al. "Walk on stars: A grid-free monte carlo method for pdes with neumann boundary conditions." arXiv preprint arXiv:2302.11815 (2023), hereinafter referred to as “Sawhney”, in view of Wang et al. "All-frequency rendering of dynamic, spatially-varying reflectance." In ACM SIGGRAPH Asia 2009 papers, pp. 1-10. 2009, hereinafter referred to as “Wang”, and in further view of Xu et al. "Warped-area reparameterization of differential path integrals." ACM Transactions on Graphics (TOG) 42, no. 6 (2023): 1-18, hereinafter referred to as “Xu”.
Regarding claim 1, Sawhney discloses a method, comprising:
generating an initialized cone representative of a region of a scene, wherein the initialized cone is defined by a central ray and an initialized angle from the central ray (Section 5.1.1-5.1.2; we adopt the spatialized normal cone hierarchy (SNCH)…Each node of a SNHC stores not only an axis-aligned bounding box (AABB), but also a normal cone. The cone axis is the average normal of all triangles in the node, and the cone half angle θ is the maximum angle between the axis and any triangle normal (Figure 12). Normal cones can be assembled during BVH construction…For each node N in this traversal we build a view cone rooted at x. The cone's axis points toward the center of N, and its half-angle tightly bounds the AABB (Figure 12)…);
tracing the central ray to determine visibility information for the central ray (Algorithm 1 and Section 5; Q.3 ray intersection queries against 𝜕ΩN…use standard closest point and ray intersection algorithms);
based on the visibility information for the central ray, detecting a closest silhouette boundary to the central ray (Fig. 11-12 and Section 5.1.2; finding the closest silhouette edge…perform a silhouette query, we traverse the SNCH in depth-first order (Algorithm 2)…Section E, Algorithm 2; Output: The closest point to x on the visibility silhouette of 𝜕ΩN…Algorithm 3; Output: Distance from x to closest point on visibility silhouette)
Sawhney does not disclose a device, detecting a closest silhouette boundary to the central ray by angle, and outputting an indication of the central ray and an angle of the closest silhouette boundary to the central ray as parameters of a continuous visibility gradient.
In the same art of representing and evaluating scene visibility relative to a query direction, Wang discloses a device (Pg. 133:7, Section 8; We implemented our rendering algorithm on a PC with an Intel CoreTM2 Duo 3.2 CPU, 4GB memory, and an nVidia Geforce 8800Ultra graphics card…):
detecting a closest silhouette boundary to the central ray by angle (Pg. 133:2, Section 1; For visibility V, we use a per-vertex spherical signed sitance function (SSDF) which encodes the angular distance to the closest visibility boundary at each direction. Pg. 133:5, Section 5.1; The function's sign encodes whether the direction I is occluded (negative) or not (positive), and its value encodes angular distance, θd, to the nearest direction t on the shadow boundary…Eq. (23)…); and
outputting an indication of the central ray and an angle of the closest silhouette boundary to the central ray as parameters (Fig. 5 and Pg 133:5, Section 5-5.2; ghost-free interpolation….Visibility, Vx(i), is a spatially varying binary spherical function indicating whether the direction I to a distance light source is occluded or not, at each surface point x…We map binary visibility to an SSDF, Vd(i)…Eq. (23) and Eq. (24)…)
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 Sawhney’s computer-based closest silhouette query using Wang’s GPU rendering system and signed angular-distance representation. Sawhney and Wang are in the same art of representing and evaluating scene visibility relative to a query direction, and Wang’s representation provides a compact, processor-suitable way to encode directional visibility information. One skilled in the art would have recognized that applying Wang’s angular representation and graphics processing architecture to Sawhney’s returned silhouette boundary information would facilitate automated GPU execution, ghost-free interpolation of high-frequency visibility, and efficient integration of the visibility information with lighting and reflectance calculations to provide a stable and efficient rendering processes (Wang Abstract and Section 1).
Sawhney in view of Wang does not disclose of a continuous visibility gradient.
In the same art of processing visibility boundaries of a scene, Xu discloses a continuous visibility gradient (Fig. 6-7 and Section 3.2.2; specify a continuous velocity field v in the interior of the domain Bwae such that v agrees with 𝒗𝜕 on the domain boundary 𝜕Bwae…Eq. (14)…Appendix B Eq. (58)-(62)…where the gradient ∇w of the kernel w and the divergence ∇· 𝒗dis𝐾 of the discontinuous velocity 𝒗dis𝐾 can both be computed using automatic differentiation).
It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to process Sawhney and Wang’s combined angular visibility-boundary representation using Xu’s continuous field formulation so that the boundary information could supply a differentiable visibility result. Xu explains that visibility discontinuous can produce severely biased derivative estimates and that its formulation can efficiently be estimated using advanced Monte Carlo rendering methods without expressly sampling discontinuity boundaries (Xu Abstract and Section 1). The modification would have yielded predictable results by converting the detected angular visibility boundary into a continuous derivative representation suitable for differentiable and inverse-rendering operations.
Regarding claim 3, Sawhney in view of Wang in further view of Xu discloses the method of claim 1, and Sawhney further discloses wherein the visibility information indicates a visible geometry (Fig. 8, Fig. 10 and Section 4.4.2; let r be the minimum of the distance dDirichlet to the Dirichlet boundary, and the distance d silhouette to the closest point on the visibility silhouette of 𝜕ΩN (Section 5). The connected component of B(xk, r) ∩ Ω containing xk then defines a star-shaped region St(xk, r). Figure 8 shows several examples…only visible parts of the Neumann boundary 𝜕ΩN…).
Sawhney, Wang, and Xu are combined for the reasons set forth above with respect to claim 1.
Regarding claim 4, Sawhney in view of Wang in further view of Xu discloses the method of claim 1, and Sawhney further discloses wherein the central ray is traced using a bounding volume hierarchy generated for the scene (Fig. 11-12 and Section 5-5.1.1; All queries use a bounding volume hierarchy (BVH) to achieve amortize sublinear scaling…).
Sawhney, Wang, and Xu are combined for the reasons set forth above with respect to claim 1.
Regarding claim 5, Sawhney in view of Wang in further view of Xu discloses the method of claim 1, and Sawhney further discloses wherein the central ray is traced by traversing the bounding volume hierarchy in a depth-first order prioritizing bounding volumes (Section 5.1.2; To perform a silhouette query, we traverse the SNCH in depth-first order (Algorithm 2). For each node N in this traversal we build a view cone rooted at x…Section E, Algorithm 3; …Traverse closer subtree first…).
Sawhney does not disclose, doing so with a smallest angular distance.
In the same art of representing and evaluating scene visibility relative to a query direction, Wang teaches measuring closeness to a visibility boundary as an angular distance from a query direction, and discloses a smallest angular distance (Section 5.1; encodes angular distance, θd, to the nearest direction t on the shadow boundary…Eq. (23)…).
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 Sawhney’s depth first order traversal to prioritize the eligible bounding volume having the smaller angular distance according to Wang. Sawhney prioritizes the spatially closer subtree when searching for the closest silhouette, while Wang teaches measuring closeness to a visibility boundary as an angular distance from a query direction. Angular distance would have been a predictable alternative measure of proximity for selecting the next subtree, and doing so would allow later nodes to be rejected more efficiently and avoid exhaustive edge examination.
Regarding claim 7, Sawhney in view of Wang in further view of Xu discloses the method of claim 1, and Sawhney further discloses wherein the closest silhouette boundary is detected by traversing a bounding volume hierarchy generated for the scene (Section 5; Q.2 closest silhouette point queries to 𝜕ΩN…All queries use a bounding volume hierarchy…spatialized normal cone hierarchy (SNCH)…Section 5.1.2; Closest Silhouette Point Traversal…).
Sawhney, Wang, and Xu are combined for the reasons set forth above with respect to claim 1.
Regarding claim 9, Sawhney in view of Wang in further view of Xu discloses the method of claim 7, and Sawhney further discloses wherein the closest silhouette boundary is detected by traversing the bounding volume hierarchy in a depth-first order (Section 5.1.2; To perform a silhouette query, we traverse the SNCH in depth-first order (Algorithm 2). For each node N in this traversal we build a view cone rooted at x…Section E, Algorithm 3; …Traverse closer subtree first…)
Sawhney does not disclose traversing the bounding volume hierarchy in a depth-first order by angle
In the same art of representing and evaluating scene visibility relative to a query direction, Wang discloses that proximity to the nearest visibility boundary can be represented as an angular distance from the query direction, and discloses by angle (Section 5.1; encodes angular distance, θd, to the nearest direction t on the shadow boundary…Eq. (23)…).
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 Wang’s angular distance measure in Sawhney’s disclosed depth first BVH traversal. Replacing Sawhney’s spatial distance comparison with Wang’s angular comparison would allow preserving Sawhney’s traversal structure while adapting the traversal order to the directional cone query. This modification would prioritize nodes more likely to contain the closest boundary in angular space, reducing unnecessary traversal and edge testing.
Regarding claim 10, Sawhney in view of Wang in further view of Xu discloses the method of claim 9, and Sawhney further discloses wherein during the traversal, a subtree within the bounding volume hierarchy is culled when the minimal (Algorithm 3; if dminT > r then return r ⊲ Ignores nodes outside search radius…ignore nodes outside search radius… ⊲ Traverse closer subtree first…)
Sawhney does not disclose the minimal angle for a bounding volume represented by the subtree is larger than an angle.
In the same art of representing and evaluating scene visibility relative to a query direction, Wang supplies an angular measure of the distance from a query direction to the nearest visibility boundary, and discloses the minimal angle for a bounding volume represented by the subtree is larger than an angle (Section 5.1; encodes angular distance, θd, to the nearest direction t on the shadow boundary…Eq. (23)…)
It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to apply Wang’s angular distance representation to Sawhney’s subtree culling procedure in order to use the corresponding angular quantities in Sawhney’s comparison during closest boundary search. A subtree whose minimum possible angle exceeds the latest closest boundary angle cannot improve the current result, and therefore may be culled to avoid unnecessary node and primitive tests.
Regarding claim 11, Sawhney in view of Wang in further view of Xu discloses the method of claim 9, and Sawhney further discloses wherein during the traversal, only a subset of all edges of a bounding volume are tested (Fig. 12 and Section 5.1.2; If this test fails, all triangles in N must be front- or back-facing relative to the query point, and it can be skipped…Section E, Algorithm 3. Examiner's note: skips an entire node when the view and normal cones cannot contain orthogonal direction, only edges in reached leaf nodes are tested)
Sawhney, Wang, and Xu are combined for the reasons set forth above with respect to claim 1.
Regarding claim 14, Sawhney in view of Wang in further view of Xu discloses the method of claim 1, and Sawhney further discloses wherein the closest silhouette boundary is detected using a single query (Algorithm 2; SilhouetteDistanceNeumann(x, rmax)…Input: A query point 𝑥 ∈ R3 & a radius r around x to search in. Output: The closest point to x on the visibility silhouette of 𝜕ΩN. 1: retrun DISTCLOSESTSILHOUETTE(snch.root, x rmax)).
Sawhney, Wang, and Xu are combined for the reasons set forth above with respect to claim 1.
Regarding claim 15, Sawhney in view of Wang in further view of Xu discloses the method of claim 1, and Sawhney further discloses wherein the closest silhouette boundary is detected using multiple queries (Section 5.1.2 and Algorithm 3).
Sawhney, Wang, and Xu are combined for the reasons set forth above with respect to claim 1.
Regarding claim 17, Sawhney in view of Wang in further view of Xu discloses the method of claim 1, but Sawhney does not disclose wherein the central ray and the angle of the closest silhouette boundary to the central ray defines a cone-shaped region in the scene having no visibility changes to the visibility information.
In the same art of representing and evaluating scene visibility relative to a query direction, Wang discloses wherein the central ray and the angle of the closest silhouette boundary to the central ray defines a cone-shaped region in the scene having no visibility changes to the visibility information (Fig. 6 and Pg. 133:5, Section 5.1; Visibility, Vx(i), is a spatially-varying binary spherical function indicating whether the direction I to a distance light source is occluded or not, at each surface point x…binary visibility to an SSDF, Vd(i)…encodes whether the direction I is occluded (negative) or not (positive), and its value encodes angular distance, θd, to the nearest t on the shadow boundary…Eq. (23)…).
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 Wang’s signed angular distance to identify the visibility region of Sawhney’s central query direction. Doing so allows the system to reuse a common visibility classification within the region and further provide ghost-free interpolation of high-frequency visibility (Wang Abstract, Section 1, and Section 5.1)
Regarding claim 19, Sawhney in view of Wang in further view of Xu discloses the method of claim 1, but Sawhney does not disclose wherein the method is an inverse rendering method.
In the same art of processing visibility boundaries of a scene, Xu discloses wherein the method is an inverse rendering method (Fig. 12-17 and Section 6.2-6.4).
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 Sawhney and Wang’s combined continuous visibility gradient information in Xu’s inverse rendering process. Applying the resulting gradients to inverse rendering would allow scene parameters, such as geometry, to be iteratively adjusted to reduce a different between rendered and targeted observations.
Regarding claim 20, Sawhney in view of Wang in further view of Xu discloses the method of claim 1, but Sawhney does not disclose wherein the parameters are output for providing differentiable visibility
In the same art of processing visibility boundaries of a scene, Xu discloses wherein the parameters are output for providing differentiable visibility (Section 3.1.1; the discontinuity curves ΔBe consist of material points p whose spatial representations x = X(p, θ) are jump discontinuities of mutual visibility V(x ↔ y) (with fixed y. Section 3.2, Eq. (11). Examiner's note: uses visibility boundary information to calculate the derivative contribution required for differentiable rendering and output)
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 Xu’s differentiable rendering process into Sawhney and Wang’s combined system. Incorporating the angular boundary parameters would allow the renderer to include visibility changes and handle geometry-induced visibility discontinuities to avoid viased derivative estimates (Xu Section 1).
Regarding claim 21, Sawhney in view of Wang in further view of Xu discloses the method of claim 20, but Sawhney does not disclose wherein the parameters define boundaries of a region adjacent to a visibility discontinuity in the scene.
In the same art of processing visibility boundaries of a scene, Xu discloses wherein the parameters define boundaries of a region adjacent to a visibility discontinuity in the scene (Fig. 5 and Section 3.2.1; Let Bewa be a subset of the reference surface Be containing all material points p whose spatial representations x = X(p, θ) are visible to y…Eq. (12)…the boundary 𝜕Bwae of Bewa consists of material points p whose spatial representations x = X(p, θ) belong to one of the following categories…C.2 Silhouette of the emitter…C.3 Visibility boundaries from occlusion).
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 the closest boundary parameters of Sawhney and Wang’s combined system with Xu’s visible region adjacent to a visibility discontinuity. Doing so avoids the need to expressly sample the discontinuity curve (Xu Abstract and Section 1).
Regarding claim 22, Sawhney in view of Wang in further view of Xu discloses the method of claim 21, but Sawhney does not disclose wherein in the region a new corrective term is added to account for a missing visibility gradient.
In the same art of processing visibility boundaries of a scene, Xu discloses wherein in the region a new corrective term is added to account for a missing visibility gradient (Section 3.1; Eq. (4)…The boundary component in Eq. (4)…is a line integral that is unique to differentiable rendering…Section 3.1; rewrite boundary integrals as interior ones using the divergence theorem…smoothly interpolating 𝒗𝜕 from the boundary 𝜕Bwae to the interior Bwae…rewrite the full material-form differential integral (4) as one interior integral…Eq. (11). Examiner's note: The divergence term ∇· (𝐹dir𝒗)] is added to the interior derivative to account for the visibility-boundary contribution that would be missing, therefore is the claimed new corrective term).
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 Xu’s divergence-based corrective contribution into the visibility gradient calculation of Sawhney and Wang’s combined system. The angular boundary detected by Sawhney and Wang’s combined system identifies a visibility discontinuity whose derivative contribution would not be captured by an ordinary derivative taken only within continuous regions. Xu applies the divergence theorem to move the boundary contribution into an interior integral and includes the resulting divergence term in rendering calculations. Adding such contribution would account for missing visibility boundary derivative, and further reduce the severe bias associated with visibility changes (Xu Section 1).
Regarding claim 24, Sawhney in view of Wang in further view of Xu discloses the method of claim 20, but Sawhney does not disclose wherein the central ray and the angle of the closest silhouette boundary to the central ray defines a cone-shaped region that provides an area of known support for integrating.
In the same art of representing and evaluating scene visibility relative to a query direction, Wang discloses wherein the central ray and the angle of the closest silhouette boundary to the central ray defines a cone-shaped region that provides an area of known support for integrating (Section 5.1; encodes angular distance, θd, to the nearest direction t on the shadow boundary…Eq. (23)…Fig. 6; t is the nearest visibility boundary to the lobe center p. Section 5.3; integrating the lobe over the intersection of its hemispherical support…with the visibility hemisphere (whose boundary passes through t nearest to p)…Eq. (28)-(3). Examiner's note: the central lobe direction p and the closest boundary angle θd establish angular limits δ0= π/2−θd of the integration region).
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 Wang’s closest boundary angle as the support boundary for integrating visibility dependent rendering quantities of Sawhney’s methods. The closest boundary angle supplies a known angular limit for the supported integration region, and using that region would allow efficient evaluation of visibility, lighting, and reflectange of fast integrated products (Wang Section 1 and 5.3).
Regarding claim 25, Sawhney in view of Wang in further view of Xu discloses the method of claim 1, but Sawhney does not disclose wherein the parameters are output for forward rendering.
In the same art of processing visibility boundaries of a scene, Xu discloses wherein the parameters are output for forward rendering (Section 4.1.1; Path integrals for forward rendering…).
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 Xu’s visibility parameters available during forward rendering into Sawhney and Wang’s combined system. Using the boundary parameters during the forward pass allows the renderer to compute the visibility dependent image value and retain corresponding boundary information for subsequent calculations, and further avoid redundant visibility evaluation.
Regarding claim 26, Sawhney in view of Wang in further view of Xu discloses the method of claim 1, but Sawhney does not disclose wherein the closest silhouette boundary indicates a closest
point on an occluding shadow edge which is used during the forward rendering to approximate a soft shadow.
In the same art of representing and evaluating scene visibility relative to a query direction, Wang discloses wherein the closest silhouette boundary indicates a closest point on an occluding shadow edge which is used during the forward rendering to approximate a soft shadow (Section 5.1; encodes angular distance, θd, to the nearest direction t on the shadow boundary…Fig. 6; t is the nearest visibility boundary to the lobe center p. b) Reconstructed visibility function around p…Section 5.3; integrating the lobe over the intersection of its hemispherical support…with the visibility hemisphere (whose boundary passes through t nearest to p). Section 6; The radius parameter actually models a spherical light source: increasing r increases the softness of shadow boundaries)
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 Wang’s closest shadow-boundary point during forward rendering to approximate the visibility of an extended light source into Sawhney’s methods. Doing so provides an efficient approximation of the portion of an extended light that remains visible, resulting in a soft shadow (Wang Section 6).
Regarding claim 28, claim 28 is the system claim (Wang Pg. 133:7, Section 8; We implemented our rendering algorithm on a PC with an Intel CoreTM2 Duo 3.2 CPU, 4GB memory, and an nVidia Geforce 8800Ultra graphics card…) of method claim 1, and is accordingly rejected using substantially similar rationale as to that which is set for with respect to claim 1.
Regarding claim 29, claim 29 has similar limitations as of claim 25, except it is a system claim (Wang Pg. 133:7, Section 8; We implemented our rendering algorithm on a PC with an Intel CoreTM2 Duo 3.2 CPU, 4GB memory, and an nVidia Geforce 8800Ultra graphics card…), and is therefore rejected under the same rationale as claim 25.
Regarding claim 30, claim 30 has similar limitations as of claim 26, except it is a system claim (Wang Pg. 133:7, Section 8; We implemented our rendering algorithm on a PC with an Intel CoreTM2 Duo 3.2 CPU, 4GB memory, and an nVidia Geforce 8800Ultra graphics card…), and is therefore rejected under the same rationale as claim 26.
Regarding claim 31, claim 31 is the CRM claim (Wang Pg. 133:7, Section 8; We implemented our rendering algorithm on a PC with an Intel CoreTM2 Duo 3.2 CPU, 4GB memory, and an nVidia Geforce 8800Ultra graphics card…) of method claim 1, and is accordingly rejected using substantially similar rationale as to that which is set for with respect to claim 1.
Regarding claim 32, claim 32 has similar limitations as of claim 25, except it is a CRM claim (Wang Pg. 133:7, Section 8; We implemented our rendering algorithm on a PC with an Intel CoreTM2 Duo 3.2 CPU, 4GB memory, and an nVidia Geforce 8800Ultra graphics card…), and is therefore rejected under the same rationale as claim #.
Claim(s) 2, 6, 8, and 18 is/are rejected under 35 U.S.C. 103 as being unpatentable over Sawhney, in view of Wang, in further view of Xu, and in further view of Shkurko (US 20230206544 A1).
Regarding claim 2, Sawhney in view of Wang in further view of Xu discloses the method of claim 1, but does not disclose wherein the initialized angle is predefined.
In the same art of iterative cone-based traversal of bounding volume acceleration structures, Shkurko discloses wherein the initialized angle is predefined (Par. 0014; An intersection test is then performed to determine whether there is both angular overlap and distance between the frustum circle and the bounding circle…Par. 0021; A circle of a sphere that bounds the frustum where it intersects the hemisphere is identified as the "frustum circle"…Par. 0023-0024; the frustum 302 is a frustum of a rectangular pyramid extending from the camera 306, but the 3D volume that forms the basis of the frustum 302 can be implemented using any of a variety of volumes, such as a circular or elliptical cone, a triangular pyramid (tetrahedron), a pentagonal pyramid, a hexagonal pyramid…With the frustum 302 determined, at block 206 the intersection test stage 128 identifies a central ray 308 that extends from the camera 306 through the center of the frustum 302…Examiner's note:API defines the frustum shape, and is then projected as a frustum circle bounding its angular extend on the hemisphere. Thus, the selected frustum shape has a predetermined angular extent before BVH traversal).
It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to initialize Shawney, Wang, and Xu’s combined system having a cone-based hierarchy query using Shkurko’s predetermined shape and to associate the nodes with Shkurko’s angular bounding circle. Doing so allows a coherent group of directions to be evaluated against the hierarchy using a common angular representation, yielding predictable results in more efficient BVH intersection testing, with fewer dot-product operations and a transformation cost amortized over the node tests (Shkurko Par. 0015).
Regarding claim 6, Sawhney in view of Wang in further view of Xu discloses the method of claim 1, and Sawhney further discloses wherein the closest silhouette boundary is detected by reducing the (Section E, Algorithm 3; Input: A spatialized normal cone hierarchy T, a query point 𝑥 ∈ R3, a radius r around x to search in…Output: Distance from x to closest point on visibility silhouette…if dminT > r then return r ⊲ Ignores nodes outside search radius…ignore nodes outside search radius… ⊲ Traverse closer subtree first…)
Sawhney does not disclose reducing the initialized angle over one or more steps.
In the same art of iterative cone-based traversal of bounding volume acceleration structures, Shkurko discloses reducing the initialized angle over one or more steps (Par. 0021; A central ray that extends from the camera point through the center of the frustum is identified and a transform matrix that converts the frustum from the Cartesian world coordinate system to a spherical coordinate system with the central ray defining the Z-axis is determined. A hemisphere centered on the Z-axis/central ray having a specified radius is defined. Par. 0034-0040; an iteration of the projection hemisphere-based intersection test of block 406 is initiated with the selection of the next node of a BVH or other ray traversal acceleration structure at block 404…to determine whether the frustum 302 and the bounding volume of the current node intersect, at block 508 the intersection test stage 128 evaluates the frustum circle 324 and the BV circle to determine if (1) there is angular overlap…).
It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to update the initialized angular search bound as Sawhney, Wang, and Xu’s combined system detects closer silhouette candidates. Doing so allows nodes outside the updated cone to be rejected earlier, reducing unnecessary traversal and intersection testing, and improving overall BVH testing efficiency (Shkurko Par. 0015)
Regarding claim 8, Sawhney in view of Wang in further view of Xu discloses the method of claim 7, but does not disclose wherein the bounding volume hierarchy includes a minimal angle from the central ray to each of a plurality of bounding volumes in the bounding volume hierarchy.
In the same art of iterative cone-based traversal of bounding volume acceleration structures, Shkurko discloses wherein the bounding volume hierarchy includes a minimal angle from the central ray to each of a plurality of bounding volumes in the bounding volume hierarchy (Par. 0021; A central ray that extends from the camera point through the center of the frustum is identified and a transform matrix that converts the frustum from the Cartesian world coordinate system to a spherical coordinate system with the central ray defining the Z-axis is determined…Thereafter, a BVH or other bounding volume acceleration structure can be traversed, and for each node to be tested, the bounding volume at that node is then bound by a sphere…the bounding sphere is then projected onto the hemisphere using the transform matrix to determine a projected circle of a sphere (the “bounding circle”) representing the bounding sphere in the spherical coordinate system). Fig. 5-6 and Par. 0036; the bounding sphere is transformed to “frustum space” using the world-to-frustum transform 312. The radius “r” and angle (“theta”) between the sphere center to the Z-axis are then identified, with theta =arc cos(z/r), the center. The transformed bounding sphere is then projected onto the surface of the projection hemisphere 320, and the resulting circle of a sphere (or circle of a hemisphere) on the surface of the projection hemisphere 320 is designated as the BV circle for the bounding volume of the current node of the BVH. Examiner's note; angular position is theta =arc cos(z/r) of the node's bounding sphere relative to the central ray, and projecting the sphere as a bounding circle representing its angular extent. The angular position and angular extent of each bounding circle define the minimum angle from the central ray to that bounding volume)
It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to augment Sawhney, Wang, and Xu’s combined BVH traversal with Shkurko’s angular position and angular extent for each bounding volume. Doing so would allow the traversal to perform Shkurko’s angular overlap determination and reject bounding volumes that cannot intersect the ray cone, yielding predictable results of fewer dot product operations and amortization of the transform cost (Shkurko Par. 0015).
Regarding claim 18, Sawhney in view of Wang in further view of Xu discloses the method of claim 7, but does not disclose wherein the central ray and the angle of the closest silhouette boundary to the central ray defines a cone-shaped region in the scene having no ray-geometry intersections.
In the same art of iterative cone-based traversal of bounding volume acceleration structures, Shkurko discloses wherein the central ray and the angle of the closest silhouette boundary to the central ray defines a cone-shaped region in the scene having no ray-geometry intersections (Par. 0023. Par. 0037-0040).
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 Shkurko’s frustum miss determination to identify an angular region in which further ray geometry testing is unnecessary. Doing so allows the system to avoid performing individual ray geometry tests and improve intersection testing efficiency (Shkurko Par. 0015, 0021)
Claim(s) 12 and 27 is/are rejected under 35 U.S.C. 103 as being unpatentable over Sawhney, in view of Wang, in further view of Xu, and in further view of Clark et al. (US 20210407172 A1), hereinafter referred to as “Clark”.
Regarding claim 12, Sawhney in view of Wang in view of Xu discloses the method of claim 11, but does not disclose wherein between 1 and 6 edges of the bounding volume are tested.
In the same art of ray intersection testing using bounding volumes and silhouette edges, Clark discloses wherein between 1 and 6 edges of the bounding volume are tested (Fig. 2 and Par. 0130; box 201 is an axis-aligned bounding box (AABB)…silhouette edges are determined from a line of sight along the direction of the ray vector…positioned at the ray's origin and directed along the direction of the ray. In other words, the silhouette edges, labelled 1 to 6 in FIG. 2, are the edges of the 2D projection box 201…has either four or six silhouette edges depending upon the perspective form which the box is viewed…Par. 0140-141; tests for each of the silhouette edges labelled 1 to 6 in FIG. 6…to perform a ray intersection test for box 201, an edge test is performed for each silhouette edge 1 to 6 to determine what side of each of those edges the ray passes on, where each edge test involves performing an evaluation to determine whether the condition specified in equation (11) is true for the two planes forming the edge under test).
It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to limit the bounding volume edge tests as taught by Clark, in Sawhney, Wang, and Xu’s combined traversal. Edges that do not form part of the projected silhouette cannot define whether the ray passes inside or outside the projected box and therefore does not need to be tested. Therefore, reducing the number of required calculations decreases ray-tracing latency, increases performance, and reduces power consumption (Clark Par. 0110).
Regarding claim 27, Sawhney in view of Wang in view of Xu discloses the method of claim 25, but does not disclose wherein the closest silhouette boundary is used to determine which neighbors can be reused at minimal cost during the forward rendering.
In the same art of ray intersection testing using bounding volumes and silhouette edges, Clark discloses wherein the closest silhouette boundary is used to determine which neighbors can be reused at minimal cost during the forward rendering (Fig. 4 and Par. 0110; rays can be grouped together into a bundle of rays, and an intersection test can be performed for the bundle of rays. In a situation where the intersection testing results would be the same for all of the rays within a bundle then, rather than performing all of the intersection tests for each of the individual rays within the bundle, it may be possible to perform fewer intersection tests (e.g. one or two intersection tests) for the bundle of rays as a whole, and to use the result(s) of the test(s) to infer the intersection testing results for all of the rays in the bundle…bundle intersection test(s). Par. 0151; a bundle of rays is received at the tester module 307. The bundle of rays is to be tested for intersection with a box…).
It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to apply Clark’s ray bundle reuse to that neighborhood in Sawhney, Wang, and Xu’s combined closest visibility boundary system. Doing so reduces the number of intersection tests and memory access (Clark Par. 0110; separate intersection tests for individual rays within at least some of the bundles may be avoided…performing intersection testing on bundles of rays can reduce the overall number of intersection tests that are performed and can result in fewer accesses to memory for fetching ray data…By reducing (compared to the prior art) the number of calculations that need to be performed to determine whether the rays of a bundle intersect with a box, the intersection testing results can be determined faster, i.e. the latency of the ray tracing system can be reduced (or in other words, the performance of the ray tracing system can be increased), and the power consumption of the ray tracing system can be reduced).
Claim(s) 13 is/are rejected under 35 U.S.C. 103 as being unpatentable over Sawhney, in view of Wang, in further view of Xu, and in further view of Maillot et al. (US 20090051684 A1), hereinafter referred to as "Maillot”.
Regarding claim 13, Sawhney in view of Wang in view of Xu discloses the method of claim 9, but does not disclose wherein a geometric normal of a surface intersected by the central ray is used as a clipping plane for the traversal.
In the same art of computer generated 3D graphics using viewing rays, Maillot discloses wherein a geometric normal of a surface intersected by the central ray (Fig. 7C-7D and Par. 0067-0068; FIG. 7C illustrates a ray 704 that is generated into the frustum…FIG. 7D illustrates a normal 708 to the intersection point Q(x,y,z) 706) is used as a clipping plane for the traversal (Par. 0068-0069; A normal to the intersection point Q(x,y,z) 706 is calculated which determines the tangent plane 710 which is perpendicular to the normal 708 and intersects the intersection point 706 Q(x,y,z)…Portions falling outside of the world-space brush are identified and essentially clipped by applying the tangent plane normal's inverse transformation to the vertices. Par. 0094-005; …the process checks 1526 to see if there is an intersection. If there is no intersection, then the process proceeds to the looping 1516 operation and loops to the next triangle).
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 traversal of the combined system of Sawhney, Wang, and Xu to use a tangent clipping plane defined at the central-ray intersection by the intersected surface's normal, as taught by Maillot. Incorporating this local plane-based clipping into the traversal allows eliminating geometry inconsistent with the visible surface before performing additional intersection processing, therefore reducing unnecessary processing when rejecting invalid primitives.
Claim(s) 16 is/are rejected under 35 U.S.C. 103 as being unpatentable over Sawhney, in view of Wang, in further view of Xu, and in further view of Clarberg (US 20150084981 A1).
Regarding claim 16, Sawhney in view of Wang in view of Xu discloses the method of claim 1, but does not disclose wherein the closest silhouette boundary is a virtual silhouette boundary introduced by non-manifold self-intersections between interpenetrating triangles.
In the same art of identifying visibility discontinues in rendering, Clarberg discloses wherein the closest silhouette boundary is a virtual silhouette boundary introduced by non-manifold self-intersections between interpenetrating triangles (Par. 0023-0024; perform high-quality antialiasing of all visible triangle edges, as well as all visible intersections of interpenetrating triangles…each triangle edge that requires high-quality anti-aliasing is marked as a discontinuity edge…detect silhouette edges…).
It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to treat Clarberg’s visible intersections between interpenetrating triangles as additional silhouette discontinuity boundaries in the Sawhney, Wang, and Xu’s combined visibility processing method. Doing so would prevent the renderer from omitting visible discontinuities that do not correspond to ordinary manifold mesh edges, providing high-quality antialiasing (Clarberg Par. 0023-0024).
Claim(s) 23 is/are rejected under 35 U.S.C. 103 as being unpatentable over Sawhney, in view of Wang, in further view of Xu, and in further view of Bangaru et al. "Unbiased warped-area sampling for differentiable rendering." ACM Transactions on Graphics (TOG) 39, no. 6 (2020): 1-18, hereinafter referred to as “Bangaru”.
Regarding claim 23, Sawhney in view of Wang in view of Xu discloses the method of claim 22, but does not disclose wherein the gradient is linearly ramped to give a constant divergence field and lower variance.
In the same art of differentiable rendering techniques, Bangaru discloses wherein the gradient is linearly ramped (Fig. 16 and Pg. 245:11, Section 4.2; Our intersection derivative �𝜃𝜔 is no longer uniform but exhibits approximately linear variation…Appendix D; we construct an approximate estimator by finding a linear approximation for g * V…) to give a constant divergence field (Pg. 245:11, Section 4.2; we can directly compute the divergence of the warp if we know the slope of the underlying linear surface…Appendix D) and lower variance (Pg. 245:11, Section 4.2; we use this approximate estimator as a control variate to reduce the variance of our unbiased estimator…Appendix D).
It would have been obvious to a person of ordinary skill in the art, before the effective filing date of the claimed invention, to apply Bangaru’s linear control estimator technique to Sawhney, Wang, and Xu’s combined divergence-based visibility correction. Bangaru expressly states that this control variate reduces the variance of the unbiased estimator, with Figure 16 showing that the linear approximation can almost completely remove the variance in examples. Therefore, the modification would predictably improve the stability and convergence of corrective visibility gradient estimates.
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
/CHENG YUAN TSENG/Primary Examiner, Art Unit 2615
