DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim(s) 1-7, 9-16, 18-20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Meister et al. (NPL, “Parallel Locally-Ordered Clustering for Bounding Volume Hierarchy Construction”, 2018) in view of Lombardi et al. (US Publication Number 2022/0245910 A1).
(1) regarding claim 1:
As shown in fig. 1, Meister disclosed a method (page 2, Fig. 1. GPU path tracing of the Power Plant scene (12.8M triangles) using a BVH constructed by our method (left). Visualization of the number of ray intersection operations for our method (middle) and the state-of-the-art ATRBVH method [5] (right).) comprising:
identifying one or more nodes for which a nearest neighbor search is to be performed (page 2, section 3 BVH Construction, para. [0001], To identify suitable nearest neighbors for the clustering, we use sorting based on the Morton codes with local exploration of the neighborhood in the sorted sequence i.e. the nearest neighbors for the cluster is identified);
applying data characterizing the one or more nodes to a neural network model to obtain outputs identifying one or more nearest neighbors (fig. 3, page 3, note that Fig. 3. Illustration of the proposed algorithm. In each iteration, we merge mutually corresponding nearest neighbors (green nodes connected by a dotted line). New clusters and not merged clusters (red nodes) enter the next iteration. The process is repeated until only one cluster remains); and
generating a portion of a bounding volume hierarchy ("BVH") based on the one or more nearest neighbors (page 3, section 3, para. [0001], note that an algorithm using parallel locally-ordered clustering (PLOC) for BVH construction. The algorithm employs two main ideas: (1) perform locally-ordered clustering on large numbers of clusters in parallel. (2) To identify suitable nearest neighbors for the clustering, we use sorting based on the Morton codes with local exploration of the neighborhood in the sorted sequence).
Meister disclosed most of the subject matter as described as above except for specifically teaching a neural network model.
However, Lombardi disclosed a neural network model (para. [0037], note that the machine learning model may include a neural network (NN), a convolutional neural network (CNN), a generative adversarial neural network (GAN), a deep reinforcement learning (DRL) algorithm, a deep recurrent neural network (DRNN), a classic machine learning algorithm such as random forest, k-nearest neighbor (KNN) algorithm, k-means clustering algorithms).
At the time of filing for the invention, it would have been obvious to a person of ordinary skilled in the art to teach a neural network model. The suggestion/motivation for doing so would have been in order to enable MVP model 400 to rebuild the BVH on a per-frame basis, thus handling dynamic scenes, and provides efficient intersection tests (para. [0051]). Therefore, it would have been obvious to combine Meister with Lombardi to obtain the invention as specified in claim 1.
(2) regarding claim 2:
Meister further disclosed the method of claim 1, wherein the identifying comprises identifying one or more nodes of the BVH that have no parent in the BVH (page 4, para. [0001], note that to determine the indices of the new interior nodes in the node buffer, we perform a parallel prefix scan on the new clusters. We determine the actual node indices by adding the prefix scan value to the node counter).
(3) regarding claim 3:
Meister disclosed most of the subject matter as described as above except for specifically teaching wherein the neural network model comprises a multi-layer perceptron.
However, Lombardi disclosed wherein the neural network model comprises a multi-layer perceptron (para. [0040], note that multilayer perceptron (MLP) routine 446 identifies vertex positions of a guide mesh 456, which is used to compute a base transformation 461 for each primitive).
At the time of filing for the invention, it would have been obvious to a person of ordinary skilled in the art to teach a neural network model. The suggestion/motivation for doing so would have been in order to enables MVP model 400 to rebuild the BVH on a per-frame basis, thus handling dynamic scenes, and provides efficient intersection tests (para. [0051]). Therefore, it would have been obvious to combine Meister with Lombardi to obtain the invention as specified in claim 3.
(4) regarding claim 4:
Meister further disclosed the method of claim 1, wherein the data characterizing the one or more nodes to the neural network model comprises one or more bounding volumes for the nodes (page 5, 3.6, para. [0002], note that the resulting BVH contains exactly one triangle per leaf. Collapsing some subtrees to leaf nodes may decrease the total SAH cost).
(5) regarding claim 5:
Meister disclosed most of the subject matter as described as above except for specifically teaching wherein the data includes maxima and minima for each axis for the bounding volumes.
However, Lombardi disclosed wherein the data includes maxima and minima for each axis for the bounding volumes (para. [0051], note that this enables MVP model 400 to rebuild the BVH on a per-frame basis, thus handling dynamic scenes, and provides efficient intersection tests. Ray marching tool 448 computes and stores volume primitives 415 that each ray intersects, wherein the ray may be defined by vector 410. The intersection of volume primitive 415 with a ray defined by vector 410 determines a domain of integration (t.sub.min, t.sub.max)).
At the time of filing for the invention, it would have been obvious to a person of ordinary skilled in the art to teach wherein the data includes maxima and minima for each axis for the bounding volumes. The suggestion/motivation for doing so would have been in order to enables MVP model 400 to rebuild the BVH on a per-frame basis, thus handling dynamic scenes, and provides efficient intersection tests (para. [0051]). Therefore, it would have been obvious to combine Meister with Lombardi to obtain the invention as specified in claim 5.
(6) regarding claim 6:
Meister disclosed most of the subject matter as described as above except for specifically teaching wherein the maxima and minima are quantized.
However, Lombardi disclosed wherein the maxima and minima are quantized (para. [0051], note that while marching along a ray between t.sub.min and t.sub.max, ray marching tool 448 checks each sample against the ray-specific list of intersected primitives).
At the time of filing for the invention, it would have been obvious to a person of ordinary skilled in the art to teach wherein the data includes maxima and minima for each axis for the bounding volumes. The suggestion/motivation for doing so would have been in order to enables MVP model 400 to rebuild the BVH on a per-frame basis, thus handling dynamic scenes, and provides efficient intersection tests (para. [0051]). Therefore, it would have been obvious to combine Meister with Lombardi to obtain the invention as specified in claim 6.
(7) regarding claim 7:
Meister disclosed most of the subject matter as described as above except for specifically teaching wherein the maxima and minima are in fixed point format.
However, Lombardi disclosed wherein the maxima and minima are in fixed point format (para. [0053], note that the alpha value associated with the pixel is set as Ap=T (t.sub.max)).
At the time of filing for the invention, it would have been obvious to a person of ordinary skilled in the art to teach wherein the maxima and minima are in fixed point format. The suggestion/motivation for doing so would have been in order to enables MVP model 400 to rebuild the BVH on a per-frame basis, thus handling dynamic scenes, and provides efficient intersection tests (para. [0051]). Therefore, it would have been obvious to combine Meister with Lombardi to obtain the invention as specified in claim 7.
(8) regarding claim 9:
Meister further disclosed the method of claim 1, wherein the neural network provides outputs for multiple levels of the BVH for a single set of inputs (page 2, 3.1, para. [0001], note that the agglomerative clustering algorithm starts with the scene triangles trivially forming n clusters with a single tri angle per cluster (n is the number of triangles). These clusters correspond to the leaves of the BVH. Then the algorithm builds the higher levels of the BVH by merging the clusters from the lower levels).
The proposed rejection of claims 1-7 and 9, renders obvious the steps of the system of claims 10-16 and 18 and the non-transitory computer-readable medium claims 19-20 because these steps occur in the operation of the proposed rejection as discussed above. Thus, the arguments similar to that presented above for claims 1-7, 9 are equally applicable to claims 10-16, 18-20.
Allowable Subject Matter
Claims 8 and 17 are objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims. The following is a statement of reasons for the indication of allowable subject matter: the prior arts made of record do not teach “wherein a number of bits in values of the fixed point format are dependent on a level in the BVH of the nodes or are based on ranges of the maxima and minima”, as recited in claims 8 and 17.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Hughes et al. (US Publication Number 2021/0149680 A1) disclosed data processing and more particularly to data processing via a general-purpose graphics processing unit.
Chen et al. (WO 2024/081177 A1) disclosed methods for enhancing a distribution of graph feature embeddings in an embedding space to improve discrimination of graph features by a graph neural network (GNN) that may include receiving a dataset comprising graph data associated with a graph, calculating a distance between a first set of node embeddings and a second set of node embeddings, determining a measure of uniformity for the dataset, determining a plurality of groups of node embeddings, determining a measure of alignment for the plurality of groups of node embeddings, generating a set of graph features based on the measure of uniformity, the measure of alignment, and the distance, and training the GNN based on the set of graph features to provide a trained GNN. Systems and computer program products are also disclosed.
Any inquiry concerning this communication or earlier communication from the examiner should be directed to Hilina K Demeter whose telephone number is (571) 270-1676.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, King Y. Poon could be reached at (571) 270- 0728. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about PAIR system, see http://pari-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/HILINA K DEMETER/Primary Examiner, Art Unit 2617
Prosecution Insights