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 Claims
Claims 1, 9 and 16 are amended.
Claim Rejections - 35 USC § 112
The following is a quotation of the first paragraph of 35 U.S.C. 112(a):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
The following is a quotation of the first paragraph of pre-AIA 35 U.S.C. 112:
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.
Claims 1, 9 are rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), first paragraph, as failing to comply with the written description requirement. The claim(s) contains subject matter which was not described in the specification in such a way as to reasonably convey to one skilled in the relevant art that the inventor or a joint inventor, or for applications subject to pre-AIA 35 U.S.C. 112, the inventor(s), at the time the application was filed, had possession of the claimed invention. Claim 1 recites “applying depth supervision and discarding depth points that fail to meet a threshold to improve depth perception”. The term “depth perception” is not disclosed anywhere within the Specification. Applicant mentions support for the amendment is found in paragraphs [0061]-[0066] of the present specification. Paragraphs [0061]-[0066] and the present specification as a whole fail to disclose improving the “depth perception”. Claims 9 and 16 limitations are similar in scope and functions performed by the computer implemented method of claim 1. Therefore, claims 9 and 16 limitations are also rejected with the same rationale as regarding claim 1.
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) 1-2, 4-5, 7-10, 12-13, 15-18 and 20 are rejected under 35 U.S.C. 103 as being unpatentable over YANG (No. US-20210150799-A1 “Yang”) in view of WANG (Wang, C., Sun, J., Liu, L., Wu, C., Shen, Z., Wu, D., ... & Zhang, L. (2023, August). Digging into depth priors for outdoor neural radiance fields. In Proceedings of the 31st ACM International Conference on Multimedia (pp. 1221-1230). (Year: 2023) “Wang”) and in further view of OST (Ost, J., Mannan, F., Thuerey, N., Knodt, J., & Heide, F. (2021). Neural scene graphs for dynamic scenes. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 2856-2865). (Year: 2021) “Ost”) in view of AROUDJ (No. US-20230260200-A1 “Aroudj”).
Regarding claim 1, Yang teaches “A computer-implemented method for synthesizing an image, comprising:” (methods; 0016); (synthetic and real images; Para 0011);
“capturing data from a scene;” (can use sensor data, e.g., camera and LiDAR data, collected during a single pass; Para 0017);
“decomposing a captured scene into static objects, dynamic objects and sky;” (Vehicles can be considered rigid dynamic objects; Para 0031); (Therefore, in the image 204, the sky 214 is automatically filled in; Para 0036);
“generating bounding boxes for the dynamic objects;” (high-quality 3D bounding box annotations to accumulate LiDAR points from multiple scans for each object of interest, e.g., vehicle; Para 0031);
“blending the sky into a merged version of the dynamic objects and the static objects; and” (after application of the surfel-GAN model, the gaps, holes, and other spaces that are missing in the first image 202 are automatically filled in. Therefore, in the image 204, the sky 214 is automatically filled in, possibly with simulated clouds, and the tops of trees are automatically filled in with simulated tree data. In addition, other objects; Para 0036);
However, while Yang fails to teach the limitation “applying depth supervision and discarding depth points that fail to meet a threshold to improve depth perception”.
Wang teaches “applying depth supervision and discarding depth points that fail to meet a threshold to improve depth perception;” (adding depth supervision improves quality significantly; Pg.6, Fig 2 Desc); (filtering out predictions farther than the threshold; Pg.8, Para 4.5.3 Threshold Clipping); ((b) depth supervision is an option for dense view, i.e., the depth supervision is necessary if the corresponding application needs the employed NeRF to have a better geometry; Pg.8, Para 5 Conclusion);
Wang discloses depth priors as supervision in ana outdoor NeRF and reporting that depth supervision to improve synthesized view quality and scene geometry. Wang also discloses providing Depth filtering process using threshold clipping. Depth predictions that are farther than the selected threshold are filtered out before use. This teaches the claimed subject matter of discarding depth points that fail the threshold.
It would be obvious to a person skilled in the art to modify Yang’s scene reconstruction and image processing to employ the threshold filter depth supervision taught by Wang because it would improve the reconstructed geometry and threshold clipping would allow depth information to be used while discarding depth points outside the range.
The motivation for the above is to have improved depth and geometry quality for easier scenes reconstruction with depth filtering techniques.
However, Yang and Wang fail to teach “simulating motion of the dynamic objects as static with movement of the bounding boxes;” “merging the dynamic objects and the static objects according to density and color of sample points;” and “synthesizing an image from volume rendered rays”.
Ost teaches “simulating motion of the dynamic objects as static with movement of the bounding boxes;” (The global location po of a dynamic object changes between frames... local three-dimensional cartesian coordinate frames Fo, fixed and aligned with an object’s pose... scaling with the inverse length of the bounding box size; Pg. 2859, Para Object Coordinate Frame); (translating the ray to an object local frame and applying an efficient AABB-ray intersection test; Pg.2860, Para Ray-box Intersection);
“merging the dynamic objects and the static objects according to density and color of sample points;” (a color and volumetric density are computed, and we calculate the pixel color by applying volumetric integration along the ray; Pg.2860, Para 4.1); (transmitted color c(r(ti)) and volumetric density σ(r(ti)) at each intersection point is predicted from the respective radiance fields in the static background node Fθbckg or dynamic node Fθc; Pg.2860, Para Volumetric Rendering);
Ost discloses an object’s local coordinate system which showcases object representation. The representation of dynamic object is defined in its static/local object coordinate system and the object’s global pose, the location of associated bounding box local frame, is transformed between frames. Ost also discloses density and color of points being computed can calculated with integration along the ray. This showcases the claimed subject matter of dynamic and static objects being integrated or merged based on color and density. While not explicitly sating “merging” the static and dynamic objects enter a common volumetric rendering calculation and the density and color are used in the resulting pixel, this is functionally a merging of the representations. It would be obvious to a person skilled in the art to utilize Yang’s LiDAR, camera information and reconstructed driving environment and Ost’s neural volumetric scene representation because both Yang and Ost representations of driving scenes contains static surroundings and dynamic objects. This would allow Yang’s reconstructed dynamic vehicles to be represented and repositioned within the environment while retaining the object volumetric representations.
The motivation for the above is to have accurate information about movement with dynamic and static objects for better representation of scene.
However, Yang, Wang and Ost fail to teach “synthesizing an image from volume rendered rays”.
However, Aroudj teaches “synthesizing an image from volume rendered rays.” (using multi-view volumetric inverse differentiable rendering (IDR) and SGD (lines 4 - 18). ... input image pixels using importance sampling (line 5); cast a ray for each selected pixel into the scene (line 6); Para 0052);
Yang, Wang and Ost and Aroudj are analogous art as they are related to scene optimization for 3D reconstruction and simulated sensor data.
The motivation for the above is to have accurate and easier scenes based on the simulation of objects in a scene and rendering rays of an image in the scene.
Therefore, it would have been obvious for an ordinary skilled person in the art before the effective filing date of claimed invention to have modified Yang by applying depth supervision and discarding depth points that fail to meet a threshold to improve depth perception as taught by Wang and by simulating motion of the dynamic objects as static with movement of the bounding boxes and merging the dynamic objects and the static objects according to density and color of sample points as taught by Ost and by synthesizing an image from volume rendered rays as taught by Aroudj.
Regarding claim 2, while Yang, Wang and Ost do not teach the limitations of claim 2, Aroudj teaches “The method of claim 1, wherein simulating motion of the dynamic objects includes encoding the dynamic objects using neural radiance field (NeRF).” (3D reconstruction may include approaches based on implicit representation (such as NeRF); Para 0004);
Aroudj discloses NeRF that is used for 3D reconstruction which relates to encoding dynamic objects.
The motivation for the above is to have accurate and efficient scenes based on the simulation of objects in a scene using NeRF.
Therefore, it would have been obvious for an ordinary skilled person in the art before the effective filing date of claimed invention to have modified Yang, Wang and Ost by simulating motion of the dynamic objects includes encoding the dynamic objects using neural radiance field (NeRF) as taught by Aroudj.
Regarding claim 4, while Yang, Wang and Ost do not teach the limitations of claim 2, Aroudj teaches “The method of claim 1, wherein merging the dynamic objects and the static objects includes rendering a scene when an intersection between a bounding box of a dynamic object and a ray view direction occurs.” (The multi-view images may depict the scene from a plurality of distinct viewing directions. ... for each of the set of multi-view images, a plurality of corners associated with a scene axis-aligned bounding box associated with the scene. ... The view ray may be represented based on the scene model; Para 0045);
Aroudj discloses multi-view images that have both viewing direction of a ray and bounding box that is associated with the scene. This relates to our interaction of the bounding box and the ray view directions.
The motivation for the above is to have more accurate and efficient scenes rendering.
Therefore, it would have been obvious for an ordinary skilled person in the art before the effective filing date of claimed invention to have modified Yang, Wang and Ost by merging the dynamic objects and the static objects includes rendering a scene when an intersection between a bounding box of a dynamic object and a ray view direction occurs as taught by Aroudj.
Regarding claim 5, while Yang, Wang and Ost do not teach the limitations of claim 2, Aroudj teaches “The method of claim 1, wherein blending the sky into the merged version includes alpha blending.” (rendered using ray marching and alpha blending; Para 0042);
The motivation for the above is to have more accurate and efficient scenes with alpha blending.
Therefore, it would have been obvious for an ordinary skilled person in the art before the effective filing date of claimed invention to have modified Yang, Wang and Ost by blending the sky into the merged version includes alpha blending as taught by Aroudj.
Regarding claim 7, Yang teaches “The method of claim 1, further comprising training a self-driving vehicle using synthesized images from the volume rendered rays.” (synthesized sensor data that can also be used for training dataset augmentation for deep neural networks; Para 0062); (sensor data for autonomous vehicle; Para 0002);
Yang discloses synthesized data that can used for training which relates to the claimed subject matter of synthesized images that is used to train the self-driving vehicle.
The motivation for the above is to have more accurate scenes and images used to better train the model.
Regarding claim 8, Yang teaches “The method of claim 7, wherein the training occurs when the self-driving vehicle is operating.” (obtaining training data for performing autonomous vehicle simulations; Para 0010); (can be easily extendable to new scenes that are driven by a self-autonomous vehicle; Para 0011);
The motivation for the above is to have more efficient training of the system when the autonomous vehicle is operating.
Regarding claim 9, Yang teaches “A system for synthesizing an image, comprising:” (a system; Para 0006);
“a hardware processor; and” (one or more processors; Para 0066);
“a memory that stores a computer program which, when executed by the hardware processor, causes the hardware processor to:” (for storing computer program instructions and data include all forms of non-volatile memory; Para 0069);
Claim 9 is directed to a system and its limitations are similar in scope and functions performed by the computer implemented method of claim 1. Therefore, claim 9 limitations are also rejected with the same rationale as regarding claim 1.
Claim 10 is directed to a system and its limitations are similar in scope and functions performed by the computer implemented method of claim 2. Therefore, claim 10 limitations are also rejected with the same rationale as regarding claim 2.
Claim 12 is directed to a system and its limitations are similar in scope and functions performed by the computer implemented method of claim 4. Therefore, claim 12 limitations are also rejected with the same rationale as regarding claim 4.
Claim 13 is directed to a system and its limitations are similar in scope and functions performed by the computer implemented method of claim 5. Therefore, claim 13 limitations are also rejected with the same rationale as regarding claim 5.
Claim 15 is directed to a system and its limitations are similar in scope and functions performed by the computer implemented method of claim 7. Therefore, claim 15 limitations are also rejected with the same rationale as regarding claim 7.
Regarding claim 16, Yang teaches A computer program product for synthesizing an image, the computer program product comprising a non-transitory computer readable storage medium having program instructions embodied therewith, the program instructions executable by a hardware processor to cause the hardware processor to:” (A computer program (which may also be referred to or described as a program, software, a software application, a module, a software module, a script, or code) can be written in any form of programming language; Para 0065); (Computer-readable media suitable for storing computer program instructions and data include all forms of non-volatile memory, media and memory devices; Para 0069); (one or more modules of computer program instructions encoded on a tangible non-transitory program; Para 0063);
Claim 16 is directed to a computer program and its limitations are similar in scope and functions performed by the computer implemented method of claim 1. Therefore, claim 16 limitations are also rejected with the same rationale as regarding claim 1.
Claim 17 is directed to a computer program and its limitations are similar in scope and functions performed by the computer implemented method of claim 2. Therefore, claim 17 limitations are also rejected with the same rationale as regarding claim 2.
Claim 18 is directed to a computer program and its limitations are similar in scope and functions performed by the computer implemented method of claim 4. Therefore, claim 18 limitations are also rejected with the same rationale as regarding claim 4.
Claim 20 is directed to a computer program and its limitations are similar in scope and functions performed by the computer implemented method of claim 7. Therefore, claim 20 limitations are also rejected with the same rationale as regarding claim 7.
Claim(s) 3 and 11 are rejected under 35 U.S.C. 103 as being unpatentable over YANG in view of WANG and in further view of OST in view of AROUDJ and in further view of DEL ROCCO (Del Rocco, J., Bourke, P. D., Patterson, C. B., & Kider, J. T., "Real-time spectral radiance estimation of hemispherical clear skies with machine learned regression models.", 2020, Solar Energy, 204, pp. 48-63. (Year: 2020), “Rocco”).
Regarding claim 3, while Yang, Wang, Ost and Aroudj fail to teach all of claim 3, Rocco teaches “The method of claim 1, further comprising modeling the sky with a sphere radiance model.” (Fig. 2 showcases a spherical radiance model);
The motivation for the above is to have a more efficient and accurate modeling of the sky within the scene.
Therefore, it would have been obvious for an ordinary skilled person in the art before the effective filing date of claimed invention to have modified Yang, Wang, Ost and Aroudj by modeling the sky with a sphere radiance model as taught by Rocco.
Claim 11 is directed to a system and its limitations are similar in scope and functions performed by the computer implemented method of claim 3. Therefore, claim 11 limitations are also rejected with the same rationale as regarding claim 3.
Claim(s) 6, 14, and 19 are rejected under 35 U.S.C. 103 as being unpatentable over YANG in view of WANG and in further view of OST in view of AROUDJ and in further view of MCREYNOLDS (McReynolds, T., & Blythe, D., "Advanced graphics programming using OpenGL", 2005, Elsevier, pp.185-209 (Year: 2005), “McReynolds”).
Regarding claim 6, while Yang, Wang, Ost and Aroudj fail to teach all of claim 6, McReynolds teaches the method of claim 5, wherein a rendered color in an image includes c = a*cvehicle&background + (1-a)*csky, where a is accumulated weight, cvehicle&background is a merged color between a vehicle and background and csky represents the color of the sky wherein a sky region is segmented out by enforcing (1-a) to be close to 1 at the sky region with a loss.” (The resulting equation is Cnew = αfCf +(1−αf)Cb; Pg. 187);
McReynolds discloses an equation used for alpha bending which relates with the claimed equation. The claimed equation is the alpha bending equation. Since it is an equation, the variables from the disclosed equation can be substituted by the variables of the claimed subject matter.
The motivation for the above is to have a more accurate calculation of the rendered color of the sky within the scene.
Therefore, it would have been obvious for an ordinary skilled person in the art before the effective filing date of claimed invention to have modified Yang, Wang, Ost and Aroudj by the rendered color in an image includes c = a*cvehicle&background + (1-a)*csky, where a is accumulated weight, cvehicle&background is a merged color between a vehicle and background and csky represents the color of the sky wherein a sky region is segmented out by enforcing (1-a) to be close to 1 at the sky region with a loss as taught by McReynolds.
Claim 14 is directed to a system and its limitations are similar in scope and functions performed by the computer implemented method of claim 6. Therefore, claim 14 limitations are also rejected with the same rationale as regarding claim 6.
Claim 19 is directed to a computer program and its limitations are similar in scope and functions performed by the computer implemented method of claim 6. Therefore, claim 19 limitations are also rejected with the same rationale as regarding claim 6.
Response to Arguments
Applicant’s arguments, see pg.9, filed 06/29/2026, with respect to Claims 16-20 have been fully considered and are persuasive. The Rejection of 04/03/2026 has been withdrawn.
Due to the amendment of claim 16 to include the term “non-transitory”, claim 16 has overcome the pervious rejection. All claims 17-20, which depend on claim 16, also overcome the rejection by virtue of dependency.
Applicant’s arguments, see pg.9, filed 06/29/2026, with respect to Claims 1, have been fully considered and are persuasive. The Rejection of 04/03/2026 has been withdrawn.
The previous rejection under 35 USC § 112(b) has been withdrawn due to amendment of the claims. Applicant should make note of new rejection under 35 USC § 112(a), see the stated reasons above.
Applicant’s arguments, see pg.9-19, filed 06/29/2026, with respect to the rejection(s) of claim(s) 1, 9 and under 35 USC § 103 have been fully considered and are persuasive. Therefore, the rejection has been withdrawn. However, upon further consideration, a new ground(s) of rejection is made in view of Yang in view of Wang in view of Ost and in further view of Aroudj.
Applicant argues that Yang does not disclose independently selecting, filtering, weighting, or modifying individual depth samples after surfel generation. Rather, the depth information contributes collectively to the estimated surfel parameters. Also, that Aroudj does mention depth as a standalone feature, but does not teach or suggest at least removing or discarding depth data. Also, that the Examiner has failed to establish a prima facie case of unpatentability because the rejection does not identify teachings or suggestions for each and every claim limitation as originally filed. And that Aroudj fails to cure the deficiencies of Yang.
Examiner replies that the rejection does not rely on Yang in view of Arouji to teach or suggest the claimed limitations. With respective to the combination of new ground rejection, necessitated by the amendment of claim 1, Yang in view of Wang in view of Ost and in further view of Aroudj, the combination teaches or suggest the claimed subject matter.
Applicant argues that Yang does not teach or suggest at least "simulating motion of the dynamic objects as static with movement of the bounding boxes" (Office Action, p. 5). That the Examiner applies Aroudj to cure this deficiency. However, neither Yang nor Aroudj, alone or in combination, teaches or suggests the claimed element. Claim 1 recites, inter alia, "simulating motion of the dynamic objects as static with movement of the bounding boxes." This can be explained as, motion is represented through movement of the bounding boxes while the underlying dynamic objects remain static. Yang teaches moving the object, not moving the bounding box, and therefore fails to teach present claim 1. Aroudj does not disclose dynamic objects, object-specific bounding boxes, movement of bounding boxes, simulation of object motion, or maintaining dynamic objects as static representations. The AABB of Aroudj is not associated with a dynamic object and is not used to represent motion. Rather, it simply defines the outer limits of the scene being reconstructed. Therefore, the cited combination fails to teach or suggest the limitation "simulating motion of the dynamic objects as static with movement of the bounding boxes."
Examiner replies that with respective to the combination of new ground rejection, Yang in view of Wang in view of Ost and in further view of Aroudj, the combination teaches or suggest the claimed subject matter. Yang uses 3D bounding boxes in reconstructing dynamic objects and Aroudj describes a scene level AABB. Ost more expressly teaches the claimed object motion representation. That the global location of a dynamic object changes between frames while establishing a local 3D coordinate frame that is fixed and aligned to the object’s pose. Ost further teaches scaling the local coordinate frame according to the bounding box dimensions and performing ray intersections with the objects’ AABB after transforming the ray into the object’s local coordinate frame (SEE Pg.2859, Para Object Coordinate Frame). Ost therefore teaches the dynamic object as locally static while simulating motion through movement/transformation of the bounded object frame.
Applicant argues that the prior art fails to teach or suggest at least "merging the dynamic objects and the static objects according to density and color of sample points." Aroudj mentions "sampling density," but this is not taught or suggested to be in relation to the merging of objects. Therefore, the cited combination fails to teach or suggest the limitation "merging the dynamic objects and the static objects according to density and color of sample points." Accordingly, the rejection must be withdrawn.
Examiner replies that with respective to the combination of new ground rejection, Yang in view of Wang in view of Ost and in further view of Aroudj, the combination teaches or suggest the claimed subject matter. In view of Ost, Ost teaches static scene and dynamic objects nodes and during rendering, it obtains color and density values. Ost expressly teaches that the sample points map to a predicted color and density value and that the rendering integrates sampling points with the ray. Thus, the samples associated with static and dynamic objects are implemented in a common volume rendering operation according to their density and color values (SEE 2860, Para 4.1 & Volumetric Rendering). Ost therefore teaches margining the dynamic and static representation according to density or color of sample points.
Applicant argues that claim 1 is amended to recite, inter alia, "applying depth supervision and discarding depth points that fail to meet a threshold to improve depth perception". Yang fails to teach or suggest at least discarding any of the surfels, and even more so does not do so because of a threshold. Therefore, Yang fails to teach or suggest at least "applying depth supervision and discarding depth points that fail to meet a threshold to improve depth perception." Accordingly, Aroudj fails to cure the deficiencies of Yang. Aroudj teaches using depth (Aroudj § [0049], FIG. 2). Aroudj, however, does not teach or suggest at least "discarding" depth data. At most, Aroudj ignores points with low resolution from far depth but does not discard any data (Aroudj $ [0072]. Therefore, Aroudj fails to teach or suggest at least "applying depth supervision and discarding depth points that fail to meet a threshold to improve depth perception."
Examiner replies that with respective to the combination of new ground rejection, Yang in view of Wang in view of Ost and in further view of Aroudj, the combination teaches or suggest the claimed subject matter. Wang expressly teaches depth supervised NeRFs and direct depth supervision in which depth is rendered by the NeRF is compared with depth prior using supervision loss. Wang also discloses that depth supervision helps improve geometry of the NeRF. Wang also teaches depth filtering in connection to depth supervised NeRF and that depth values may be filtered. Depth predictions exceeding the threshold are excluded from the depth data used in depth supervision rendering process (SEE Pg.6 Fig 2 Desc & Pg.8, Para 4.5.3, 5). Wang therefore teaches depth supervision and discarding depth values that fail to satisfy threshold. While Wang does not explicit state the improvement of depth perception, it is suggested as depth supervision is utilized to improve geometry and better the geometry. Improving geometry corresponds to improving the depth perception as the visual ability to see the scene in 3D is a better perception.
Applicant argues that the deficiencies of Yang with respect to independent claims 1, 9, and 16 are noted above. That Aroudj does not cure the above noted deficiencies of Yang with respect to independent claims 1, 9, and 16. Accordingly, Applicant respectfully submits that independent claims 1, 9, and 16 are patentable over Yang in view of Aroudj under 35 U.S.C. § 103, whether taken alone or in combination. Claims 2-8, 10-15, and 17-20 depend from one of claims 1, 9, and 16. Thus, for at least the stated reasons, claims 1, 9, and 16 are believed to be patentable, and dependent claims 2-8, 10-15, and 17-20 are likewise believed patentable over Yang in view of Aroudj under 35 U.S.C. § 103.
Examiner replies that with respective to the combination of new ground rejection, Yang in view of Wang in view of Ost and in further view of Aroudj, the combination teaches or suggests Claim 1. Claims 9 and 16 limitations are similar in scope and functions performed by the computer implemented method of claim 1. Therefore, claims 9 and 16 limitations are also rejected with the same rationale as regarding claim 1. By at least virtue of dependency, claims 2-8, 10-15, and 17-20 depend from one of claims 1, 9, and 16 and thus the rejection is maintained under new grounds of rejection.
Applicant argues that Del Rocco fails to cure the deficiency of Yang and Aroudj. Del Rocco does not include depth in its calculation, and even explicitly excludes depth information. Therefore, Del Rocco fails to teach or suggest at least "applying depth supervision and discarding depth points that fail to meet a threshold 'to improve depth perception." McReynolds includes depth data but fails to teach or suggest at least discarding or removing the data at any time (§ 11.5 Compositing Images with Depth, p. 194). McReynolds fails to cure the deficiencies of Yang, Aroudj, and Del Rocco. McReynolds teaches using depth information (§ 11.5 Compositing Images with Depth, p. 194). However, the depth information is not discarded. In other words, all depth data is saved but only useful data is used to optimize the model, and the remaining data is stored idly. Therefore, McReynolds fails to teach or suggest at least "applying depth supervision and discarding depth points that fail to meet a threshold to improve depth perception."
Examiner replies that Del Rocco or McReynolds is not relied upon for the newly recited depth supervision and threshold filter limitation. Wang is relied upon to teach applying depth supervision while excluding depth values that fail the threshold (SEE Pg.6 Fig 2 Desc & Pg.8, Para 4.5.3, 5). Therefore, the absence of geometric depth supervision in Del Rocco and McReynolds does not establish patentability over the applied combination.
Applicant argues that Del Rocco and McReynolds do not cure the deficiencies of the claims as previously recited prior to the amendments herein. For instance, Del Rocco and McReynolds do not teach or suggest at least "simulating motion of the dynamic objects as static with movement of the bounding boxes." Del Rocco does not mention, let alone teach or suggest at least bounding boxes, and only mentions dynamic in the context of high dynamic range (HDR) imagery. Therefore, Del Rocco does not teach or suggest at least "simulating motion of the dynamic objects as static with movement of the bounding boxes."
Nor does McReynolds teach or suggest at least "simulating motion of the dynamic objects as static with movement of the bounding boxes." McReynolds mentions the term "motion" but does not generate motion and instead discusses motion blur when combining images to illustrate motion occurred instead of actually generating motion (See McReynolds § 11.7 Dissolves, p. 197). Therefore, McReynolds does not teach or suggest at least "simulating motion of the dynamic objects as static with movement of the bounding boxes."
Examiner replies that Del Rocco or McReynolds is not relied upon for teaching “simulating motion of the dynamic objects as static with movement of the bounding boxes”. Ost is relied upon to teach simulating motion of dynamic objects a static with bounded boxes movement (SEE Pg.2859, Para Object Coordinate Frame). Therefore, the absence of motion representation in Del Rocco and McReynolds does not establish patentability over the applied combination.
Applicant argues that Del Rocco and McReynolds do not teach or suggest at least "merging the dynamic objects and the static objects according to density and color of sample points." Del Rocco teaches sampling colors, but specifically does not merge colors or any other data. Therefore, Del Rocco does not teach or suggest at least "merging the dynamic objects and the static objects according to density and color of sample points."
McReynolds teaches combining data to form images but the data combined is using an alpha value which relates to transparency and not "dynamic objects and the static objects" or "density," but rather, "transparency or the coverage of the object intersecting the pixel". Therefore, McReynolds does not teach or suggest at least "merging the dynamic objects and the static objects according to density and color of sample points."
Examiner replies that Del Rocco or McReynolds is not relied upon for teaching “merging the dynamic objects and the static objects according to density and color of sample points”. Ost is relied upon to teach merging dynamic and static objects from density and color (SEE 2860, Para 4.1 & Volumetric Rendering). Therefore, the absence of density/color merging in Del Rocco and McReynolds does not establish patentability over the applied combination.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Deng, K., Liu, A., Zhu, J. Y., & Ramanan, D. (2022, June). Depth-supervised nerf: Fewer views and faster training for free. In 2022 IEEE/CVF conference on computer vision and pattern recognition (CVPR) (pp. 12872-12881). IEEE. (Year: 2022) (Deng) – Discloses rendering better images given fewer training views while training 2-3x faster. Compatible with other recently proposed NeRF methods, demonstrating that depth is a cheap and easily digestible supervisory signal. And supporting other types of depth supervision such as scanned depth sensors and RGB-D reconstruction outputs
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to BRIGITER D PROTAZI whose telephone number is (571)272-7995. The examiner can normally be reached Monday - Friday 7:30-5.
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, Said A Broome can be reached at 5712722931. 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.
/B.D.P./Examiner, Art Unit 2612
/Said Broome/Supervisory Patent Examiner, Art Unit 2612
Prosecution Insights