CTNF 18/649,088 CTNF 101738 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. Information Disclosure Statement 06-52 The information disclosure statement (IDS) was submitted on 9/4/2024, 10/10/2024, 1/22/2025, and 1/28/2025. The submission is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner. Preliminary Amendment Applicant’s preliminary Amendment filed on 5/24/2024 has been entered and made of record. Currently Pending claims: 1-21 Independent claims: 1, 9, 12, and 14 Amended claims: 1, 2, 4, 9-15, 17, and 19-21 Election/Restrictions Applicant’s election without traverse of Group I, claims 1-11 and 14-21 in the reply filed on 5/11/2026 is acknowledged. Claims 12 and 13 are withdrawn. Drawings 06-22-03 AIA The drawings are objected to as failing to comply with 37 CFR 1.84(p)(4) because reference character “ 220 ” has been used to designate both the training device in Fig. 2a and NPU in Fig. 22 . Corrected drawing sheets in compliance with 37 CFR 1.121(d) are required in reply to the Office action to avoid abandonment of the application. Any amended replacement drawing sheet should include all of the figures appearing on the immediate prior version of the sheet, even if only one figure is being amended. Each drawing sheet submitted after the filing date of an application must be labeled in the top margin as either “Replacement Sheet” or “New Sheet” pursuant to 37 CFR 1.121(d). If the changes are not accepted by the examiner, the applicant will be notified and informed of any required corrective action in the next Office action. The objection to the drawings will not be held in abeyance. Claim Objections 07-29-01 AIA Claim s 6 and 19 are objected to because of the following informalities: Claim 6, lines 6-7 , "global attention mechanisms is one of attention mechanisms" is unclear. The preceding claims do not introduce a plurality of attention mechanisms. Claim 19, lines 6-7 , "global attention mechanisms is one of attention mechanisms" is unclear. The preceding claims do not introduce a plurality of attention mechanisms . Appropriate correction is required. 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. This application includes one or more claim limitations that do not use the word “means,” but are nonetheless being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, because the claim limitation(s) uses a generic placeholder that is coupled with functional language without reciting sufficient structure to perform the recited function and the generic placeholder is not preceded by a structural modifier. Such claim limitation(s) is/are: “execution device” in claims 3 and 16. Because this/these claim limitation(s) is/are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, it/they is/are being interpreted to cover the corresponding structure described in the specification as performing the claimed function, and equivalents thereof. Regarding claim 3 and 16 , “execution device” is interpreted as a terminal device, an intelligent lock, or a data monitoring or processing device as described in ¶0025, “The execution device configured with the target model may be any one of the following devices: a terminal device, an intelligent lock, and a data monitoring and processing device”. If applicant does not intend to have this/these limitation(s) interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, applicant may: (1) amend the claim limitation(s) to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph (e.g., by reciting sufficient structure to perform the claimed function); or (2) present a sufficient showing that the claim limitation(s) recite(s) sufficient structure to perform the claimed function so as to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. 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. Regarding claims 5, 8, 18, and 21 , the phrase "and/or" renders the claim indefinite because it is unclear whether the limitation(s) following the phrase are part of the claimed invention. Examiner will interpret “and/or” as “or”. Positive Statement Regarding - 35 USC § 101 The Examiner’s 35 U.S.C. 101 analysis recognizes that the claimed subject matter is directed to a practical application of a technical solution. The claimed elements, taken as a whole, improve the functioning of machine learning prediction models by enhancing the accuracy, see ¶0009. Because the claims recite specific, claimed steps and structural elements that produce a tangible technical result, they are not directed to an abstract idea absent additional inventive concept limitations. Accordingly, the record supports a positive 101 determination for the present claims. 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. 07-23-aia AIA The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows: 1. Determining the scope and contents of the prior art. 2. Ascertaining the differences between the prior art and the claims at issue. 3. Resolving the level of ordinary skill in the pertinent art. 4. Considering objective evidence present in the application indicating obviousness or nonobviousness. 07-21-aia AIA Claim s 1-4, 7-11, 14-17, 20, and 21 are rejected under 35 U.S.C. 103 as being unpatentable over Zhou et al. 2021 (US 10,970,518) (hereafter, “Zhou”) (IDS) in view of Bhattacharyya et al. (Bhattacharyya, Prarthana, Chengjie Huang, and Krzysztof Czarnecki. "Sa-det3d: Self-attention based context-aware 3d object detection." Proceedings of the IEEE/CVF international conference on computer vision . 2021) (hereafter, “Bhattacharyya”) (IDS) . Regarding claim 1 , Zhou discloses a method of point cloud data processing, the method comprising: obtaining point cloud data ( Col. 6, lines 12-13, In some embodiments, a voxel feature learning/detection network is configured to receive raw point cloud data ) corresponding to a target environment ( Col. 5, lines 15-23, A point cloud may include a set of data points within a coordinate system … For example, a point cloud may include LiDAR information regarding objects around a vehicle, such as other vehicles, pedestrians, etc.; Col. 15, lines 5-10, One experimental setup according to one example embodiment may be based on the LiDAR specifications of the KITTI dataset. Car Detection: For this task, point clouds within the range of [−3; 1]×[−40; 40]×[0; 70.4] meters along Z, Y, X axis respectively may be considered. Examiner considers the defined range as the target environment ), wherein the point cloud data include initial information of a plurality of target points ( Col. 9, lines 64-67, Voxel 302 includes points 304, 306, and 308, which each may have associated X, Y, and Z coordinates in 3D space and one or more associated attributes, such as respective reflectances ) and are divided into a plurality of target cubes ( Col. 6, lines 15-19, As described in more detail below, the voxel feature learning/detection network is configured to determine voxel features for a plurality of voxels of the point cloud, wherein a voxel is a 3D volume of the point cloud, for example a cuboid or a rectangular volume. Examiner considers the voxels as the “target cubes” ), wherein each target cube includes S target points, S being a non-negative integer ( Col. 9, lines 23-27, In some embodiments, a fixed number of points, T, from those voxels containing more than T points may be randomly sampled to reduce the number of points to be less than or equal to T number of points. Examiner considers T to be implicitly non-negative and that the sampling outcome encompasses a plurality of cubes with T points ); generating an initial feature of each target cube based on the initial information of the S target points in each target cube ( Col. 10, lines 55-59, In some embodiments, a list of voxel features may be obtained by processing only the non-empty voxels of a point cloud, such as of point cloud 202. Each voxel feature may be uniquely associated to the spatial coordinates of a particular non-empty voxel ); updating initial features of the plurality of target cubes ( Col. 3, lines 22-25, The fully connected neural network may further transform the output of the voxel feature encoding layers and apply element-wise max-pooling to determine a voxel-wise feature for the voxel. Examiner considers transforming the features with pooling as “updating the features” ) [ based on an attention mechanism to obtain updated features of the plurality of target cubes ]; and performing a feature processing operation on the updated features of the plurality of target cubes to obtain a prediction result corresponding to the point cloud data ( Col. 5, lines 13-15, learn effective features from point clouds and predict accurate 3D bounding boxes; Col. 7, lines 49-51, Object detection bounding boxes 112 included in the 3D object detection result 110 ), wherein the prediction result represents information about at least one object in the target environment ( Col. 7, lines 49-51, Object detection bounding boxes 112 included in the 3D object detection result 110; Claim 5, bounding boxes corresponding to objects … 3D bounding boxes identified by the region proposal network. Examiner considers the bounding box to “represent” information about an object ). However, Zhou fails to disclose based on an attention mechanism to obtain updated features of the plurality of target cubes. Bhattacharyya teaches based on an attention mechanism to obtain updated features of the plurality of target cubes ( Page 4, left column, top paragraph, we introduce a variant of FSA called Deformable Self-Attention; Page 4, §3.2 Full Self-Attention Module, Our Full Self-Attention (FSA) module projects the features xi through linear layers…normalizing it with group normalization [45] and summing it with xi ). Both Zhou and Bhattacharyya are analogous to the claimed invention because both are in the field of machine-learning based object detection. It would have been obvious to a person of ordinary skill before the effective filing date of the claimed invention to incorporate the attention mechanism of Bhattacharyya into the model of Zhou. The suggestion/motivation for doing so would have been to improve detection, as suggested by Bhattacharyya at page 9, §6. Conclusions, our architecture systematically improves the performance of 3D object detectors . This method of improving Zhou was within the ordinary ability of one of ordinary skill in the art based on the teachings of Bhattacharyya. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date, to modify Zhou with the teachings of Bhattacharyya to obtain the invention as specified in claim 1. Regarding claim 2 , in which claim 1 is incorporated, Zhou discloses wherein the updating initial features of the plurality of target cubes based on the attention mechanism comprises: when a first cube meets a first preset condition ( Col. 10, lines 55-57, In some embodiments, a list of voxel features may be obtained by processing only the non-empty voxels of a point cloud ), updating the feature of each first cube in the plurality of target cubes ( Col. 3, lines 22-25, The fully connected neural network may further transform the output of the voxel feature encoding layers and apply element-wise max-pooling to determine a voxel-wise feature for the voxel. Examiner considers transforming the features with pooling as “updating the features” ) [ based on the attention mechanism ], wherein the first cube is any one of the plurality of target cubes ( Col. 9, lines 56-57, The VFE layer-1 (e.g. VFE layer 208 a ) receives voxel 302 ), the first preset condition is that the first cube is a non-empty cube ( Col. 10, lines 55-57, In some embodiments, a list of voxel features may be obtained by processing only the non-empty voxels of a point cloud ), and a feature of the non-empty cube is not preset information ( Col. 10, lines 18-21, After obtaining point-wise feature representations 314, element-wise max pooling 316 may be used across all f i associated to the voxel V (e.g. voxel 302) to get the locally aggregated voxel feature. Examiner considers the aggregated feature to be “not preset” ). However, Zhou fails to explicitly disclose based on the attention mechanism. Bhattacharyya teaches based on the attention mechanism ( Page 4, §3.2 Full Self-Attention Module, Our Full Self-Attention (FSA) module projects the features xi through linear layers…normalizing it with group normalization [45] and summing it with xi ). Both Zhou and Bhattacharyya are analogous to the claimed invention because both are in the field of machine-learning based object detection. It would have been obvious to a person of ordinary skill before the effective filing date of the claimed invention to incorporate the attention mechanism of Bhattacharyya into the model of Zhou. The suggestion/motivation for doing so would have been to improve detection, as suggested by Bhattacharyya at page 9, §6. Conclusions, our architecture systematically improves the performance of 3D object detectors . This method of improving Zhou was within the ordinary ability of one of ordinary skill in the art based on the teachings of Bhattacharyya. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date, to modify Zhou with the teachings of Bhattacharyya to obtain the invention as specified in claim 2. Regarding claim 3 , in which claim 2 is incorporated, Zhou discloses wherein the method is applied to an execution device ( Col. 20, lines 43-45, In at least some embodiments, a system and/or server that implements a portion or all of one or more of the methods and/or techniques described herein. Examiner considers the server the “execution device” ), and the execution device records the non-empty cube in the plurality of target cubes by using a hash table ( Col. 14, lines 32-35, In some embodiments, this may be done efficiently using an O(1) lookup operation using a hash table (e.g. index 906) where the voxel coordinate is used as the hash key ). Regarding claim 4 , in which claim 2 is incorporated, Zhou discloses wherein the updating initial features of the plurality of target cubes based on the attention mechanism further comprises: [ when the first cube meets a second preset condition ], updating the feature of each first cube in the plurality of target cubes ( Col. 3, lines 22-25, The fully connected neural network may further transform the output of the voxel feature encoding layers and apply element-wise max-pooling to determine a voxel-wise feature for the voxel. Examiner considers transforming the features with pooling as “updating the features” ) [ based on the attention mechanism, wherein the second preset condition is that the non-empty cube exists in a cube set, the cube set comprises the first cube and at least one second cube corresponding to the first cube, and the at least one second cube is determined from the plurality of target cubes based on the attention mechanism ]. However, Zhou fails to explicitly disclose when the first cube meets a second preset condition, updating the feature based on the attention mechanism, wherein the second preset condition is that the non-empty cube exists in a cube set, the cube set comprises the first cube and at least one second cube corresponding to the first cube, and the at least one second cube is determined from the plurality of target cubes based on the attention mechanism. Bhattacharyya teaches when the first cube meets a second preset condition ( Page 4, §3.1 Formulation, the set of pillar/ voxel /point features and their relations are denoted by a graph. The voxel being in a set is the condition ), updating the feature based on the attention mechanism ( Page 4, §3.2 Full Self-Attention Module, Our Full Self-Attention (FSA) module projects the features xi through linear layers…normalizing it with group normalization [45] and summing it with xi ), wherein the second preset condition is that the non-empty cube exists in a cube set ( Page 4, §3.1 Formulation, the set of pillar/ voxel /point features and their relations are denoted by a graph ), the cube set comprises the first cube and at least one second cube corresponding to the first cube ( Page 4, §3.1 Formulation, node set V = {x1,x2,...xn ∈ Rd}. Examiner considers any of the other nodes in the node set to be “at least one second cube” ), and the at least one second cube is determined from the plurality of target cubes based on the attention mechanism ( Page 5, §3.3 Deformable Self-Attention Module, Our primary idea is to attend to a representative subset of the original node vectors …, it is essential to make sure that the selected nodes cover the informative structures and common characteristics in 3D geometric space. Examiner considers the “at least one second cube” to be determined is any one of the nodes from the representative subset ). Both Zhou and Bhattacharyya are analogous to the claimed invention because both are in the field of machine-learning based object detection. It would have been obvious to a person of ordinary skill before the effective filing date of the claimed invention to incorporate the attention mechanism of Bhattacharyya into the model of Zhou. The suggestion/motivation for doing so would have been to improve detection, as suggested by Bhattacharyya at page 9, §6. Conclusions, our architecture systematically improves the performance of 3D object detectors . This method of improving Zhou was within the ordinary ability of one of ordinary skill in the art based on the teachings of Bhattacharyya. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date, to modify Zhou with the teachings of Bhattacharyya to obtain the invention as specified in claim 4. Regarding claim 7 , in which claim 1 is incorporated, Zhou discloses wherein the point cloud data corresponding to the target environment comprises point cloud data corresponding to a surrounding environment of a target vehicle ( Col. 15, lines 5-10, One experimental setup according to one example embodiment may be based on the LiDAR specifications of the KITTI dataset. The KITTI dataset is comprised of point cloud data collected from sensors mounted on a car ), and the performing a feature processing operation on the updated features of the plurality of target cubes to obtain a prediction result corresponding to the point cloud data comprises: performing a target detection operation on the updated feature of the target cube to obtain the prediction result corresponding to the point cloud data ( Col. 12, lines 45-50, In some embodiments, the input to the region proposal network may be a feature map provided by the Convolutional Middle Layers (e.g. a high-dimensional volumetric representation of the point cloud). An example architecture of a region proposal network is illustrated in FIG. 7, according to one example framework for car detection ), wherein the prediction result indicates a location of at least one object in the surrounding environment of the target vehicle ( Col. 7, lines 49-51, Object detection bounding boxes 112 included in the 3D object detection result 110; Claim 5, bounding boxes corresponding to objects … 3D bounding boxes identified by the region proposal network; Col. 12, line 50, car detection. Examiner considers the bounding box to “represent” information about an object, in this case a car ). Regarding claim 8 , in which claim 1 is incorporated, Zhou discloses wherein the performing a feature processing operation on the updated features of the plurality of target cubes to obtain a prediction result corresponding to the point cloud data comprises: performing a target detection operation on the updated features of the plurality of target cubes to obtain the prediction result corresponding to the point cloud data ( Col. 12, lines 45-50, In some embodiments, the input to the region proposal network may be a feature map provided by the Convolutional Middle Layers (e.g. a high-dimensional volumetric representation of the point cloud). An example architecture of a region proposal network is illustrated in FIG. 7, according to one example framework for car detection ), wherein the information about the at least one object in the target environment comprises location information of the at least one object in the target environment ( Col. 7, lines 49-51, Object detection bounding boxes 112 included in the 3D object detection result 110; Claim 5, bounding boxes corresponding to objects … 3D bounding boxes identified by the region proposal network; Col. 12, line 50, car detection. Examiner considers the bounding box to “represent” information about an object, in this case a car. Since the limitation is recited in the alternative, Examiner considers this citation to fully disclose the limitation ); performing a facial recognition operation on the updated features of the plurality of target cubes to obtain the prediction result corresponding to the point cloud data, wherein the information about the at least one object in the target environment comprises category information of the at least one object in the target environment, and the category information of the at least one object in the target environment indicates that face matching succeeds or face matching fails; or performing a posture recognition operation on the updated features of the plurality of target cubes to obtain the prediction result corresponding to the point cloud data, wherein the information about the at least one object in the target environment comprises a body feature of at least one human body in the target environment and/or a gait feature of the at least one human body in the target environment during walking Regarding claim 9 , Zhou discloses a method of training a neural network, the method comprising: obtaining point cloud data ( Col. 6, lines 12-13, In some embodiments, a voxel feature learning/detection network is configured to receive raw point cloud data ) corresponding to a target environment ( Col. 5, lines 15-23, A point cloud may include a set of data points within a coordinate system … For example, a point cloud may include LiDAR information regarding objects around a vehicle, such as other vehicles, pedestrians, etc.; Col. 15, lines 5-10, One experimental setup according to one example embodiment may be based on the LiDAR specifications of the KITTI dataset. Car Detection: For this task, point clouds within the range of [−3; 1]×[−40; 40]×[0; 70.4] meters along Z, Y, X axis respectively may be considered. Examiner considers the defined range as the target environment ), wherein the point cloud data comprises initial information of a plurality of target points ( Col. 9, lines 64-67, Voxel 302 includes points 304, 306, and 308, which each may have associated X, Y, and Z coordinates in 3D space and one or more associated attributes, such as respective reflectances ) and is divided into a plurality of target cubes ( Col. 6, lines 15-19, As described in more detail below, the voxel feature learning/detection network is configured to determine voxel features for a plurality of voxels of the point cloud, wherein a voxel is a 3D volume of the point cloud, for example a cuboid or a rectangular volume. Examiner considers the voxels as the “target cubes” ), one of which includes S target points, S being a non-negative integer ( Col. 9, lines 23-27, In some embodiments, a fixed number of points, T, from those voxels containing more than T points may be randomly sampled to reduce the number of points to be less than or equal to T number of points. Examiner considers T to be implicitly non-negative and that the sampling outcome encompasses a plurality of cubes with T points ); generating an initial feature of each target cube based on the initial information of the target point in each target cube and by using a to-be-trained model ( Col. 10, lines 55-59, In some embodiments, a list of voxel features may be obtained by processing only the non-empty voxels of a point cloud, such as of point cloud 202. Each voxel feature may be uniquely associated to the spatial coordinates of a particular non-empty voxel; Col. 13, lines 42-46, A system configured to implement a generic 3D detection network as described herein may include at least three functional blocks: (1) Voxel Feature Learning Network, (2) Convolution Middle Layers, and (3) Region Proposal Network; Examiner considers the generic 3D detection network the “to-be-trained model” ), and updating initial features of the plurality of target cubes [ based on an attention mechanism ] to obtain updated features of the plurality of target cubes ( Col. 3, lines 22-25, The fully connected neural network may further transform the output of the voxel feature encoding layers and apply element-wise max-pooling to determine a voxel-wise feature for the voxel. Examiner considers transforming the features with pooling as “updating the features” ); performing a feature processing operation based on the updated features of the plurality of target cubes and by using the to-be-trained model to obtain a prediction result corresponding to the point cloud data ( Col. 5, lines 13-15, learn effective features from point clouds and predict accurate 3D bounding boxes; Col. 7, lines 49-51, Object detection bounding boxes 112 included in the 3D object detection result 110; Col. 12, lines 45-50, In some embodiments, the input to the region proposal network may be a feature map provided by the Convolutional Middle Layers (e.g. a high-dimensional volumetric representation of the point cloud). An example architecture of a region proposal network is illustrated in FIG. 7, according to one example framework for car detection ); and training the to-be-trained model according to a target loss function, wherein the target loss function indicates a similarity between the prediction result and an expected result corresponding to the point cloud data ( Col. 13, lines 20-25, Eqn. 2; Col. 13, lines 29-30, while u i ϵ R 7 and u i* ϵ R 7 may be the regression output and ground truth for positive anchor ). However, Zhou fails to explicitly disclose based on an attention mechanism. Bhattacharyya teaches based on an attention mechanism ( Page 4, §3.2 Full Self-Attention Module, Our Full Self-Attention (FSA) module projects the features xi through linear layers…normalizing it with group normalization [45] and summing it with xi ). Both Zhou and Bhattacharyya are analogous to the claimed invention because both are in the field of machine-learning based object detection. It would have been obvious to a person of ordinary skill before the effective filing date of the claimed invention to incorporate the attention mechanism of Bhattacharyya into the model of Zhou. The suggestion/motivation for doing so would have been to improve detection, as suggested by Bhattacharyya at page 9, §6. Conclusions, our architecture systematically improves the performance of 3D object detectors . This method of improving Zhou was within the ordinary ability of one of ordinary skill in the art based on the teachings of Bhattacharyya. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date, to modify Zhou with the teachings of Bhattacharyya to obtain the invention as specified in claim 9. Regarding claim 10 , in which claim 9 is incorporated, Zhou discloses wherein the updating initial features of the plurality of target cubes based on the attention mechanism comprises: when a first cube meets a first preset condition ( Col. 10, lines 55-57, In some embodiments, a list of voxel features may be obtained by processing only the non-empty voxels of a point cloud ), updating the initial feature of each first cube in the plurality of target cubes ( Col. 3, lines 22-25, The fully connected neural network may further transform the output of the voxel feature encoding layers and apply element-wise max-pooling to determine a voxel-wise feature for the voxel. Examiner considers transforming the features with pooling as “updating the features” ) [ based on the attention mechanism ], wherein the first cube is any one of the plurality of target cubes ( Col. 9, lines 56-57, The VFE layer-1 (e.g. VFE layer 208 a ) receives voxel 302 ), the first preset condition is that the first cube is a non-empty cube ( Col. 10, lines 55-57, In some embodiments, a list of voxel features may be obtained by processing only the non-empty voxels of a point cloud ), and a feature of the non-empty cube is not preset information ( Col. 10, lines 18-21, After obtaining point-wise feature representations 314, element-wise max pooling 316 may be used across all f i associated to the voxel V (e.g. voxel 302) to get the locally aggregated voxel feature. Examiner considers the aggregated feature to be “not preset” ). However, Zhou fails to explicitly disclose based on an attention mechanism. Bhattacharyya teaches based on an attention mechanism ( Page 4, §3.2 Full Self-Attention Module, Our Full Self-Attention (FSA) module projects the features xi through linear layers…normalizing it with group normalization [45] and summing it with xi ). Both Zhou and Bhattacharyya are analogous to the claimed invention because both are in the field of machine-learning based object detection. It would have been obvious to a person of ordinary skill before the effective filing date of the claimed invention to incorporate the attention mechanism of Bhattacharyya into the model of Zhou. The suggestion/motivation for doing so would have been to improve detection, as suggested by Bhattacharyya at page 9, §6. Conclusions, our architecture systematically improves the performance of 3D object detectors . This method of improving Zhou was within the ordinary ability of one of ordinary skill in the art based on the teachings of Bhattacharyya. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date, to modify Zhou with the teachings of Bhattacharyya to obtain the invention as specified in claim 10. Regarding claim 11 , in which claim 10 is incorporated, Zhou discloses wherein the updating initial features of the plurality of target cubes ( Col. 3, lines 22-25, The fully connected neural network may further transform the output of the voxel feature encoding layers and apply element-wise max-pooling to determine a voxel-wise feature for the voxel. Examiner considers transforming the features with pooling as “updating the features” ) [ based on the attention mechanism further comprises: when the first cube meets a second preset condition, updating the feature of each first cube in the plurality of target cubes based on the attention mechanism, wherein the second preset condition is that the non-empty cube exists in a cube set, the cube set comprises the first cube and at least one second cube corresponding to the first cube, and the at least one second cube is determined from the plurality of target cubes based on the attention mechanism ]. However, Zhou fails to explicitly disclose based on the attention mechanism further comprises: when the first cube meets a second preset condition, updating the feature of each first cube in the plurality of target cubes based on the attention mechanism, wherein the second preset condition is that the non-empty cube exists in a cube set, the cube set comprises the first cube and at least one second cube corresponding to the first cube, and the at least one second cube is determined from the plurality of target cubes based on the attention mechanism. Bhattacharyya teaches based on the attention mechanism ( Page 4, §3.2 Full Self-Attention Module, Our Full Self-Attention (FSA) module projects the features xi through linear layers…normalizing it with group normalization [45] and summing it with xi ) further comprises: when the first cube meets a second preset condition ( Page 4, §3.1 Formulation, the set of pillar/ voxel /point features and their relations are denoted by a graph ), updating the feature of each first cube in the plurality of target cubes based on the attention mechanism ( Page 4, §3.2 Full Self-Attention Module, Our Full Self-Attention (FSA) module projects the features xi through linear layers…normalizing it with group normalization [45] and summing it with xi ), wherein the second preset condition is that the non-empty cube exists in a cube set ( Page 4, §3.1 Formulation, the set of pillar/ voxel /point features and their relations are denoted by a graph ), the cube set comprises the first cube and at least one second cube corresponding to the first cube ( Page 4, §3.1 Formulation, node set V = {x1,x2,...xn ∈ Rd}. Examiner considers any of the other nodes in the node set to be “at least one second cube” ), and the at least one second cube is determined from the plurality of target cubes based on the attention mechanism ( Page 5, §3.3 Deformable Self-Attention Module, Our primary idea is to attend to a representative subset of the original node vectors …, it is essential to make sure that the selected nodes cover the informative structures and common characteristics in 3D geometric space. The at least one second cube is determined is any one of the nodes from the representative subset ). Both Zhou and Bhattacharyya are analogous to the claimed invention because both are in the field of machine-learning based object detection. It would have been obvious to a person of ordinary skill before the effective filing date of the claimed invention to incorporate the attention mechanism of Bhattacharyya into the model of Zhou. The suggestion/motivation for doing so would have been to improve detection, as suggested by Bhattacharyya at page 9, §6. Conclusions, our architecture systematically improves the performance of 3D object detectors . This method of improving Zhou was within the ordinary ability of one of ordinary skill in the art based on the teachings of Bhattacharyya. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date, to modify Zhou with the teachings of Bhattacharyya to obtain the invention as specified in claim 11. Regarding claim 14 , Zhou discloses a point cloud data processing apparatus, the apparatus comprising: a memory ( Col. 20, lines 53, a main memory 1220 ); and a processing device coupled with the memory ( Col. 20, lines 52-53, includes one or more processors 1210 coupled to a main memory 1220 ), the processing device and the memory configured to: obtain point cloud data ( Col. 6, lines 12-13, In some embodiments, a voxel feature learning/detection network is configured to receive raw point cloud data ) corresponding to a target environment ( Col. 5, lines 15-23, A point cloud may include a set of data points within a coordinate system … For example, a point cloud may include LiDAR information regarding objects around a vehicle, such as other vehicles, pedestrians, etc.; Col. 15, lines 5-10, One experimental setup according to one example embodiment may be based on the LiDAR specifications of the KITTI dataset. Car Detection: For this task, point clouds within the range of [−3; 1]×[−40; 40]×[0; 70.4] meters along Z, Y, X axis respectively may be considered. Examiner considers the defined range as the target environment ), wherein the point cloud data comprises initial information of a plurality of target points ( Col. 9, lines 64-67, Voxel 302 includes points 304, 306, and 308, which each may have associated X, Y, and Z coordinates in 3D space and one or more associated attributes, such as respective reflectances ), the point cloud data is divided into a plurality of target cubes ( Col. 6, lines 15-19, As described in more detail below, the voxel feature learning/detection network is configured to determine voxel features for a plurality of voxels of the point cloud, wherein a voxel is a 3D volume of the point cloud, for example a cuboid or a rectangular volume. Examiner considers the voxels as the “target cubes” ), S target points exist in each target cube, and S is an integer greater than or equal to 0 ( Col. 9, lines 23-27, In some embodiments, a fixed number of points, T, from those voxels containing more than T points may be randomly sampled to reduce the number of points to be less than or equal to T number of points. Examiner considers T to be implicitly non-negative and that the sampling outcome encompasses a plurality of cubes with T points ); generate an initial feature of each target cube based on the initial information of the target point in each target cube ( Col. 10, lines 55-59, In some embodiments, a list of voxel features may be obtained by processing only the non-empty voxels of a point cloud, such as of point cloud 202. Each voxel feature may be uniquely associated to the spatial coordinates of a particular non-empty voxel ); update initial features of the plurality of target cubes ( Col. 3, lines 22-25, The fully connected neural network may further transform the output of the voxel feature encoding layers and apply element-wise max-pooling to determine a voxel-wise feature for the voxel. Examiner considers transforming the features with pooling as “updating the features” ) [ based on an attention mechanism to obtain updated features of the plurality of target cubes ]; and perform a feature processing operation on the updated features of the plurality of target cubes to obtain a prediction result corresponding to the point cloud data ( Col. 5, lines 13-15, learn effective features from point clouds and predict accurate 3D bounding boxes; Col. 7, lines 49-51, Object detection bounding boxes 112 included in the 3D object detection result 110 ), wherein the prediction result represents information about at least one object in the target environment ( Col. 7, lines 49-51, Object detection bounding boxes 112 included in the 3D object detection result 110; Claim 5, bounding boxes corresponding to objects … 3D bounding boxes identified by the region proposal network. Examiner considers the bounding box to “represent” information about an object ). However, Zhou fails to explicitly disclose based on an attention mechanism to obtain updated features of the plurality of target cubes. Bhattacharyya teaches based on an attention mechanism to obtain updated features of the plurality of target cubes ( Page 4, left column, top paragraph, we introduce a variant of FSA called Deformable Self-Attention; Page 4, §3.2 Full Self-Attention Module, Our Full Self-Attention (FSA) module projects the features xi through linear layers…normalizing it with group normalization [45] and summing it with xi ). Both Zhou and Bhattacharyya are analogous to the claimed invention because both are in the field of machine-learning based object detection. It would have been obvious to a person of ordinary skill before the effective filing date of the claimed invention to incorporate the attention mechanism of Bhattacharyya into the model of Zhou. The suggestion/motivation for doing so would have been to improve detection, as suggested by Bhattacharyya at page 9, §6. Conclusions, our architecture systematically improves the performance of 3D object detectors . This method of improving Zhou was within the ordinary ability of one of ordinary skill in the art based on the teachings of Bhattacharyya. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date, to modify Zhou with the teachings of Bhattacharyya to obtain the invention as specified in claim 14. Regarding claim 15 , in which claim 14 is incorporated, Zhou discloses wherein the processing device and the memory ( Col. 20, lines 52-53, includes one or more processors 1210 coupled to a main memory 1220 ) are further configured to: when a first cube meets a first preset condition ( Col. 10, lines 55-57, In some embodiments, a list of voxel features may be obtained by processing only the non-empty voxels of a point cloud ), update the feature of each first cube in the plurality of target cubes ( Col. 3, lines 22-25, The fully connected neural network may further transform the output of the voxel feature encoding layers and apply element-wise max-pooling to determine a voxel-wise feature for the voxel. Examiner considers transforming the features with pooling as “updating the features” ) [ based on the attention mechanism ], wherein the first cube is any one of the plurality of target cubes ( Col. 9, lines 56-57, The VFE layer-1 (e.g. VFE layer 208 a ) receives voxel 302 ), the first preset condition is a non-empty cube ( Col. 10, lines 55-57, In some embodiments, a list of voxel features may be obtained by processing only the non-empty voxels of a point cloud ), and a feature of the non- empty cube is not preset information ( Col. 10, lines 18-21, After obtaining point-wise feature representations 314, element-wise max pooling 316 may be used across all f i associated to the voxel V (e.g. voxel 302) to get the locally aggregated voxel feature. Examiner considers the aggregated feature to be “not preset” ). However, Zhou fails to explicitly disclose based on the attention mechanism. Bhattacharyya teaches based on the attention mechanism ( Page 4, §3.2 Full Self-Attention Module, Our Full Self-Attention (FSA) module projects the features xi through linear layers…normalizing it with group normalization [45] and summing it with xi ). Both Zhou and Bhattacharyya are analogous to the claimed invention because both are in the field of machine-learning based object detection. It would have been obvious to a person of ordinary skill before the effective filing date of the claimed invention to incorporate the attention mechanism of Bhattacharyya into the model of Zhou. The suggestion/motivation for doing so would have been to improve detection, as suggested by Bhattacharyya at page 9, §6. Conclusions, our architecture systematically improves the performance of 3D object detectors . This method of improving Zhou was within the ordinary ability of one of ordinary skill in the art based on the teachings of Bhattacharyya. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date, to modify Zhou with the teachings of Bhattacharyya to obtain the invention as specified in claim 15. Regarding claim 16 , in which claim 15 is incorporated, Zhou discloses wherein the apparatus is applied to an execution device ( Col. 20, lines 43-45, In at least some embodiments, a system and/or server that implements a portion or all of one or more of the methods and/or techniques described herein. Examiner considers the server the “execution device” ), and the execution device records the non-empty cube in the plurality of target cubes by using a hash table ( Col. 14, lines 32-35, In some embodiments, this may be done efficiently using an O(1) lookup operation using a hash table (e.g. index 906) where the voxel coordinate is used as the hash key ). Regarding claim 17 , in which claim 15 is incorporated, Zhou discloses wherein the processing device and the memory ( Col. 20, lines 52-53, includes one or more processors 1210 coupled to a main memory 1220 ) are further configured to: [ when the first cube meets a second preset condition ], update the feature of each first cube in the plurality of target cubes ( Col. 3, lines 22-25, The fully connected neural network may further transform the output of the voxel feature encoding layers and apply element-wise max-pooling to determine a voxel-wise feature for the voxel. Examiner considers transforming the features with pooling as “updating the features” ) [ based on the attention mechanism, wherein the second preset condition is that the non-empty cube exists in a cube set, the cube set comprises the first cube and at least one second cube corresponding to the first cube, and the at least one second cube is determined from the plurality of target cubes based on the attention mechanism ]. However, Zhou fails to explicitly disclose when the first cube meets a second preset condition, updating the feature based on the attention mechanism, wherein the second preset condition is that the non-empty cube exists in a cube set, the cube set comprises the first cube and at least one second cube corresponding to the first cube, and the at least one second cube is determined from the plurality of target cubes based on the attention mechanism. Bhattacharyya teaches when the first cube meets a second preset condition ( Page 4, §3.1 Formulation, the set of pillar/ voxel /point features and their relations are denoted by a graph. The voxel being in a set is the condition ), updating the feature based on the attention mechanism ( Page 4, §3.2 Full Self-Attention Module, Our Full Self-Attention (FSA) module projects the features xi through linear layers…normalizing it with group normalization [45] and summing it with xi ), wherein the second preset condition is that the non-empty cube exists in a cube set ( Page 4, §3.1 Formulation, the set of pillar/ voxel /point features and their relations are denoted by a graph ), the cube set comprises the first cube and at least one second cube corresponding to the first cube ( Page 4, §3.1 Formulation, node set V = {x1,x2,...xn ∈ Rd}. Examiner considers any of the other nodes in the node set to be “at least one second cube” ), and the at least one second cube is determined from the plurality of target cubes based on the attention mechanism ( Page 5, §3.3 Deformable Self-Attention Module, Our primary idea is to attend to a representative subset of the original node vectors …, it is essential to make sure that the selected nodes cover the informative structures and common characteristics in 3D geometric space. Examiner considers the “at least one second cube” to be determined is any one of the nodes from the representative subset ). Both Zhou and Bhattacharyya are analogous to the claimed invention because both are in the field of machine-learning based object detection. It would have been obvious to a person of ordinary skill before the effective filing date of the claimed invention to incorporate the attention mechanism of Bhattacharyya into the model of Zhou. The suggestion/motivation for doing so would have been to improve detection, as suggested by Bhattacharyya at page 9, §6. Conclusions, our architecture systematically improves the performance of 3D object detectors . This method of improving Zhou was within the ordinary ability of one of ordinary skill in the art based on the teachings of Bhattacharyya. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date, to modify Zhou with the teachings of Bhattacharyya to obtain the invention as specified in claim 17. Regarding claim 20 , in which claim 14 is incorporated, Zhou discloses wherein the point cloud data corresponding to the target environment comprises point cloud data corresponding to a surrounding environment of a target vehicle ( Col. 15, lines 5-10, One experimental setup according to one example embodiment may be based on the LiDAR specifications of the KITTI dataset. The KITTI dataset is comprised of point cloud data collected from sensors mounted on a car ); and the processing device and the memory are further configured to perform a target detection operation on the updated feature of the target cube to obtain the prediction result corresponding to the point cloud data ( Col. 12, lines 45-50, In some embodiments, the input to the region proposal network may be a feature map provided by the Convolutional Middle Layers (e.g. a high-dimensional volumetric representation of the point cloud). An example architecture of a region proposal network is illustrated in FIG. 7, according to one example framework for car detection ), wherein the prediction result indicates a location of at least one object in the surrounding environment of the target vehicle ( Col. 7, lines 49-51, Object detection bounding boxes 112 included in the 3D object detection result 110; Claim 5, bounding boxes corresponding to objects … 3D bounding boxes identified by the region proposal network; Col. 12, line 50, car detection. Examiner considers the bounding box to “represent” information about an object, in this case a car ). Regarding claim 21 , in which claim 14 is incorporated, Zhou discloses wherein the processing device and the memory ( Col. 20, lines 52-53, includes one or more processors 1210 coupled to a main memory 1220 ) are further configured to: perform a target detection operation on the updated features of the plurality of target cubes to obtain the prediction result corresponding to the point cloud data ( Col. 12, lines 45-50, In some embodiments, the input to the region proposal network may be a feature map provided by the Convolutional Middle Layers (e.g. a high-dimensional volumetric representation of the point cloud). An example architecture of a region proposal network is illustrated in FIG. 7, according to one example framework for car detection ), wherein the information about the at least one object in the target environment comprises location information of the at least one object in the target environment ( Col. 7, lines 49-51, Object detection bounding boxes 112 included in the 3D object detection result 110; Claim 5, bounding boxes corresponding to objects … 3D bounding boxes identified by the region proposal network; Col. 12, line 50, car detection. Examiner considers the bounding box to “represent” information about an object, in this case a car. Since the limitation is recited in the alternative, Examiner considers this citation to fully disclose the limitation ); perform a facial recognition operation on the updated features of the plurality of target cubes to obtain the prediction result corresponding to the point cloud data, wherein the information about the at least one object in the target environment comprises category information of the at least one object in the target environment, and the category information of the at least one object in the target environment indicates that face matching succeeds or face matching fails; or perform a posture recognition operation on the updated features of the plurality of target cubes to obtain the prediction result corresponding to the point cloud data, wherein the information about the at least one object in the target environment comprises a body feature of at least one human body in the target environment and/or a gait feature of the at least one human body in the target environment during walking . 07-22-aia AIA Claim s 5, 6, 18, and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Zhou et al. 2021 (US 10,970,518) (hereafter, “Zhou”) (IDS) in view of Bhattacharyya et al. (Bhattacharyya, Prarthana, Chengjie Huang, and Krzysztof Czarnecki. "Sa-det3d: Self-attention based context-aware 3d object detection." Proceedings of the IEEE/CVF international conference on computer vision . 2021) (hereafter, “Bhattacharyya”) (IDS) as applied to claim s 1 and 14 above, and further in view of Deng et al. (Deng, Jiajun, et al. "Voxel r-cnn: Towards high performance voxel-based 3d object detection." Proceedings of the AAAI conference on artificial intelligence . Vol. 35. No. 2. 2021) (hereafter, “Deng”) . Regarding claim 5 , Zhou in view of Bhattacharyya discloses the method according to claim 4. However, neither Zhou nor Bhattacharyya, whether considered individually or in combination, explicitly disclose wherein the at least one second cube comprises all target cubes in a preset range around the first cube, and a distance between the second cube in the preset range around the first cube and the first cube is less than or equal to a preset distance threshold; and/or the at least one second cube is obtained by sampling the plurality of target cubes by using the first cube as a sampling center. Deng teaches wherein the at least one second cube comprises all target cubes in a preset range around the first cube ( Page 3, §Voxel Query, For example, the 26-neighbor voxels of a query voxel can be easily computed by adding a triplet of offsets (∆i,∆j,∆k),∆i,∆j,∆k ∈ {−1,0,1} on the voxel indices (i,j,k). This query would include all cubes within a range of 1 of the first cube ), and a distance between the second cube in the preset range around the first cube and the first cube is less than or equal to a preset distance threshold ( Page 3, §Voxel Query, For example, the 26-neighbor voxels of a query voxel can be easily computed by adding a triplet of offsets (∆i,∆j,∆k),∆i,∆j,∆k ∈ {−1,0,1} on the voxel indices (i,j,k). Examiner considers the distance threshold to be 1. Since the limitation is recited in the alternative, Examiner considers this citation to fully disclose the limitation ); and/or the at least one second cube is obtained by sampling the plurality of target cubes by using the first cube as a sampling center. Zhou, Bhattacharyya, and Deng are analogous to the claimed invention because all are in the field of machine-learning based object detection. It would have been obvious to a person of ordinary skill before the effective filing date of the claimed invention to incorporate the cube organization of Deng into the attention mechanism of Bhattacharyya and the model of Zhou. The suggestion/motivation for doing so would have been to improve performance, as suggested by Deng at page 6, left column first paragraph, Voxel R-CNN outperforms all of the existing voxel-based models by a large margin . This method of improving Zhou was within the ordinary ability of one of ordinary skill in the art based on the teachings of Bhattacharyya and Deng. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date, to modify Zhou with the teachings of Bhattacharyya and Deng to obtain the invention as specified in claim 5. Regarding claim 6 , in which claim 4 is incorporated, Zhou discloses wherein the updating the feature of each first cube in the plurality of target cubes ( Col. 3, lines 22-25, The fully connected neural network may further transform the output of the voxel feature encoding layers and apply element-wise max-pooling to determine a voxel-wise feature for the voxel. Examiner considers transforming the features with pooling as “updating the features” ) [ based on the attention mechanism further comprises: generating features of a plurality of third cubes based on a feature of the second cube, wherein one third cube comprises a plurality of second cubes ]; and updating the feature of each first cube in the plurality of target cubes ( Col. 3, lines 22-25, The fully connected neural network may further transform the output of the voxel feature encoding layers and apply element-wise max-pooling to determine a voxel-wise feature for the voxel. Examiner considers transforming the features with pooling as “updating the features” ) [ based on the feature of each third cube and a global attention mechanism, wherein the global attention mechanism is one of attention mechanisms ]. However, Zhou fails to explicitly disclose based on a global attention mechanism, wherein the global attention mechanism is one of attention mechanisms. Bhattacharyya teaches based on a global attention mechanism ( Page 4, §3.2 Full Self-Attention Module, Our Full Self-Attention (FSA) module projects the features xi through linear layers…normalizing it with group normalization [45] and summing it with xi ), wherein the global attention mechanism is one of attention mechanisms ( Page 4, §3.2 Full Self-Attention Module, Our Full Self-Attention (FSA) module ). Both Zhou and Bhattacharyya are analogous to the claimed invention because both are in the field of machine-learning based object detection. It would have been obvious to a person of ordinary skill before the effective filing date of the claimed invention to incorporate the attention mechanism of Bhattacharyya into the model of Zhou. The suggestion/motivation for doing so would have been to improve detection, as suggested by Bhattacharyya at page 9, §6. Conclusions, our architecture systematically improves the performance of 3D object detectors . However, neither Zhou nor Bhattacharyya, whether considered individually or in combination, explicitly disclose generating features of a plurality of third cubes based on a feature of the second cube, wherein one third cube comprises a plurality of second cubes. Deng teaches wherein the updating the feature of each first cube in the plurality of target cubes based on the attention mechanism further comprises: generating features of a plurality of third cubes based on a feature of the second cube ( Page 3, §Voxel Query, For example, the 26-neighbor voxels of a query voxel can be easily computed by adding a triplet of offsets (∆i,∆j,∆k),∆i,∆j,∆k ∈ {−1,0,1} on the voxel indices (i,j,k); Page 4, right column first paragraph, group neighbor voxel features with our voxel query. The query forms a 3x3x3 cube and is considered the “third cube”. Neighboring voxels are the “second cube” ), wherein one third cube comprises a plurality of second cubes ( Page 3, §Voxel Query, For example, the 26-neighbor voxels of a query voxel can be easily computed by adding a triplet of offsets (∆i,∆j,∆k),∆i,∆j,∆k ∈ {−1,0,1} on the voxel indices (i,j,k) ). Zhou, Bhattacharyya, and Deng are analogous to the claimed invention because all are in the field of machine-learning based object detection. It would have been obvious to a person of ordinary skill before the effective filing date of the claimed invention to incorporate the cube organization of Deng into the attention mechanism of Bhattacharyya and the model of Zhou. The suggestion/motivation for doing so would have been to improve performance, as suggested by Deng at page 6, left column first paragraph, Voxel R-CNN outperforms all of the existing voxel-based models by a large margin . This method of improving Zhou was within the ordinary ability of one of ordinary skill in the art based on the teachings of Bhattacharyya and Deng. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date, to modify Zhou with the teachings of Bhattacharyya and Deng to obtain the invention as specified in claim 6. Regarding claim 18 , Zhou in view of Bhattacharyya discloses the apparatus according to claim 17. However, neither Zhou nor Bhattacharyya explicitly disclose wherein the at least one second cube comprises all target cubes in a preset range around the first cube, and a distance between the second cube in the preset range around the first cube and the first cube is less than or equal to a preset distance threshold; and/or the at least one second cube is obtained by sampling the plurality of target cubes by using the first cube as a sampling center. Deng teaches wherein the at least one second cube comprises all target cubes in a preset range around the first cube ( Page 3, §Voxel Query, For example, the 26-neighbor voxels of a query voxel can be easily computed by adding a triplet of offsets (∆i,∆j,∆k),∆i,∆j,∆k ∈ {−1,0,1} on the voxel indices (i,j,k). This query would include all cubes within a range of 1 of the first cube ), and a distance between the second cube in the preset range around the first cube and the first cube is less than or equal to a preset distance threshold ( Page 3, §Voxel Query, For example, the 26-neighbor voxels of a query voxel can be easily computed by adding a triplet of offsets (∆i,∆j,∆k),∆i,∆j,∆k ∈ {−1,0,1} on the voxel indices (i,j,k). Examiner considers the distance threshold to be 1. Since the limitation is recited in the alternative, Examiner considers this citation to fully disclose the limitation ); and/or the at least one second cube is obtained by sampling the plurality of target cubes by using the first cube as a sampling center. Zhou, Bhattacharyya, and Deng are analogous to the claimed invention because all are in the field of machine-learning based object detection. It would have been obvious to a person of ordinary skill before the effective filing date of the claimed invention to incorporate the cube organization of Deng into the attention mechanism of Bhattacharyya and the model of Zhou. The suggestion/motivation for doing so would have been to improve performance, as suggested by Deng at page 6, left column first paragraph, Voxel R-CNN outperforms all of the existing voxel-based models by a large margin . This method of improving Zhou was within the ordinary ability of one of ordinary skill in the art based on the teachings of Bhattacharyya and Deng. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date, to modify Zhou with the teachings of Bhattacharyya and Deng to obtain the invention as specified in claim 18. Regarding claim 19 , in which claim 17 is incorporated, Zhou discloses wherein the processing device and the memory ( Col. 20, lines 52-53, includes one or more processors 1210 coupled to a main memory 1220 ) are further configured to: [ generate features of a plurality of third cubes based on a feature of the second cube, wherein one third cube comprises a plurality of second cubes ]; and update the feature of each first cube in the plurality of target cubes ( Col. 3, lines 22-25, The fully connected neural network may further transform the output of the voxel feature encoding layers and apply element-wise max-pooling to determine a voxel-wise feature for the voxel. Examiner considers transforming the features with pooling as “updating the features” ) [ based on the feature of each third cube and a global attention mechanism, wherein the global attention mechanism is one of attention mechanisms ] However, Zhou fails to explicitly disclose generate features of a plurality of third cubes based on a feature of the second cube, wherein one third cube comprises a plurality of second cubes; based on a global attention mechanism, wherein the global attention mechanism is one of attention mechanisms. Bhattacharyya teaches based on a global attention mechanism ( Page 4, §3.2 Full Self-Attention Module, Our Full Self-Attention (FSA) module projects the features xi through linear layers…normalizing it with group normalization [45] and summing it with xi ), wherein the global attention mechanism is one of attention mechanisms ( Page 4, §3.2 Full Self-Attention Module, Our Full Self-Attention (FSA) module ). Both Zhou and Bhattacharyya are analogous to the claimed invention because both are in the field of machine-learning based object detection. It would have been obvious to a person of ordinary skill before the effective filing date of the claimed invention to incorporate the attention mechanism of Bhattacharyya into the model of Zhou. The suggestion/motivation for doing so would have been to improve detection, as suggested by Bhattacharyya at page 9, §6. Conclusions, our architecture systematically improves the performance of 3D object detectors . However, neither Zhou nor Bhattacharyya, whether considered individually or in combination, explicitly disclose generate features of a plurality of third cubes based on a feature of the second cube, wherein one third cube comprises a plurality of second cubes. Deng teaches generate features of a plurality of third cubes based on a feature of the second cube ( Page 3, §Voxel Query, For example, the 26-neighbor voxels of a query voxel can be easily computed by adding a triplet of offsets (∆i,∆j,∆k),∆i,∆j,∆k ∈ {−1,0,1} on the voxel indices (i,j,k); Page 4, right column first paragraph, group neighbor voxel features with our voxel query. The query forms a 3x3x3 cube and is considered the “third cube”. Neighboring voxels are the “second cube” ), wherein one third cube comprises a plurality of second cubes ( Page 3, §Voxel Query, For example, the 26-neighbor voxels of a query voxel can be easily computed by adding a triplet of offsets (∆i,∆j,∆k),∆i,∆j,∆k ∈ {−1,0,1} on the voxel indices (i,j,k) ). Zhou, Bhattacharyya, and Deng are analogous to the claimed invention because all are in the field of machine-learning based object detection. It would have been obvious to a person of ordinary skill before the effective filing date of the claimed invention to incorporate the cube organization of Deng into the attention mechanism of Bhattacharyya and the model of Zhou. The suggestion/motivation for doing so would have been to improve performance, as suggested by Deng at page 6, left column first paragraph, Voxel R-CNN outperforms all of the existing voxel-based models by a large margin . This method of improving Zhou was within the ordinary ability of one of ordinary skill in the art based on the teachings of Bhattacharyya and Deng. Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date, to modify Zhou with the teachings of Bhattacharyya and Deng to obtain the invention as specified in claim 19 . Conclusion 07-96 AIA The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. Peng et al. (US 2020/0410671) discloses a voxel based lesion detection system using an attention mechanism (¶0061, use of the recurrent attention mechanism to self-adaptively process 3D lesion voxel information). Tian et al. (Tian, Yonglin, et al. "Context-aware dynamic feature extraction for 3D object detection in point clouds." IEEE Transactions on Intelligent Transportation Systems 23.8 (2021): 10773-10785) discloses extracting local context features for object detection (Fig. 2; Page 5, §3.2 Context Feature Extraction, We consider the points within the range of 3 times the width and 3 times the length of the pillar). Liu et al. (Liu, Zhe, et al. "Tanet: Robust 3d object detection from point clouds with triple attention." Proceedings of the AAAI conference on artificial intelligence . Vol. 34. No. 07. 2020) discloses an object detection network with an attention mechanism (Fig. 3; Page 3, §3D Object Detection, the Stacked Triple Attention). Any inquiry concerning this communication or earlier communications from the examiner should be directed to XIAOMAO DING whose telephone number is (571)272-7237. The examiner can normally be reached Mon-Fri 9:00-5:00. 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, Henok Shiferaw can be reached at (571) 272-4637. 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. /XIAOMAO DING/Examiner, Art Unit 2676 /CHINEYERE WILLS-BURNS/Supervisory Patent Examiner, Art Unit 2673 Application/Control Number: 18/649,088 Page 2 Art Unit: 2676 Application/Control Number: 18/649,088 Page 3 Art Unit: 2676 Application/Control Number: 18/649,088 Page 4 Art Unit: 2676
Prosecution Insights