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 .
Response to Arguments
Applicant’s response toward rejection under 35 U.S.C. 112(b) has been fully considered and is withdrawn.
Applicant's arguments filed 19 August 2026 have been fully considered but they are not persuasive. Applicant asserts the cited prior art Yang does not teach the claimed “determining attributes to be decoded at a current iteration” of claim 1. Support for their position, they applicant considers the cited Yang ¶95 essentially lists steps of prediction tree-based geometry decoding, does not reference “attributes”, and ¶58 mostly lists components of a point cloud encoder. In response, the examiner acknowledges that ¶58 lists components of a point cloud encoder and is a typographical error. However, this does not take away from the fact the preceding paragraph Yang ¶57 describes the substance of the disclosure which teaches parts of the claimed limitation. More specifically, Yang ¶57 discloses that decoding of the prior art decodes geometry information that is input into an attribute decoder to assist decompression of a point cloud attribute. Additionally, Yang ¶95 is important in disclosing that decoding successively parses a bitstream as part of geometry reconstruction at a decoding end. In essence, ¶95 was important in teaching that the process of the decoding end decoding geometry information occurs in successive parsed iterations. When combined with Yang ¶57, it teaches for decoding iterations, the decoding end decodes point cloud attributes by using decoded input geometry information of the point cloud used as additional information. In this regard, the examiner maintains that Yang teaches the determination and iteration of the claim limitation.
Applicant then argues that Krishnan does not teach “determining attributes to be decoded at a current iteration”. However, Krishnan was not relied upon to teach the limitation as maintains that the cited prior art Yang to teach the limitation does still disclose the claimed determination. See the response above to understand how Yang teaches determining attributes to be decoded at a current iteration.
The applicant argues that the cited prior art does not teach claim 10 for the same reasons the prior art does not teach claim 1. However, the examiner disagrees and has responded to this argument. See the responses above to understand how the relied upon prior art still teaches the claimed invention.
The applicant argues that the combination of prior art fails to teach “determining attributes to be encoded at a current iteration” of claim 11. In support of the applicant’s position, they assert that cited Yang ¶94 lists steps of geometry coding and does not disclose the recited feature of claim 11. They also point out that Yang ¶87 discloses attributed information as transform color information but considers this to be different from the claimed attributes to be encoded. In response, the portion of Yang ¶87 relied upon to teach the limitation relates to a point cloud encoder “during attribute encoding” which first transforms “color information (i.e., attribute information)”. This part of Yang teaches that point cloud encoding encodes attributes by first transforming attribute information. Yang ¶94 does discloses geometry coding steps, but that does not take away from the portion of Yang describes the geometry coding as through successive iterations. When combining Yand ¶87 and 94, the prior art describes the transform of attribute information during attribute encoding as part of point cloud encoding during successive iterations. This teaching describes attribute determination for encoding at current iterations, which is similar to what is being claimed. for this reason, the examiner maintains that Yang teaches the determination and iteration of the claim limitation.
The applicant argues claims 2-9 and 12-20 are allowable for the same reasons why they consider claim 1 allowable. However, the examiner disagrees and has responded to this argument. See the responses above to understand how the relied upon prior art still teaches the claimed invention.
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.
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.
Claim(s) 1,10-11 rejected under 35 U.S.C. 103 as being unpatentable over Lodhi; Muhammad Asad et al. (US 20240078715 A1) in view of YANG; Fuzheng et al. (US 20260032261 A1) in view of KRISHNAN; Madhu Peringassery et al. (US 20240080446 A1)
Regarding claim 1, Lodhi teaches,
A method of iteratively decoding point cloud data at an octree level, (¶64 and fig. 13, “decoding an encoded bitstream including compressed data representing a point cloud” including data representing a point cloud and compressed based on a “tree structure such as an octree” as depicted in fig. 13) comprising:
obtaining already-decoded attributes of a current level; (¶64 and fig. 13, “data is obtained from the encoded bitstream for a current node of the tree structure” at 1310 depicted in fig. 13)
predicting probability distributions of attributes (¶64 and fig. 13, “occupancy symbol distribution for the current node is predicted” at 1330 depicted in fig. 13) not decoded in the current level, (¶64,58, and fig. 13, occupancy symbol distribution for the current node is predicted based on feature information from “one or more ancestor nodes of the current node” such as a parent level)
wherein the predicting (¶64 and fig. 13, “occupancy symbol distribution for the current node is predicted”) is based on the already-decoded attributes of the current level (¶64 and fig. 13, “feature information from one or more available neighboring nodes” at 1330, where the neighboring nodes have already been iterated through the iteratively repeated “1310 through 1350” up to the current node) and previous levels; (¶64 and fig. 13, occupancy symbol distribution based on feature information “from one or more ancestor nodes of the current node” at 1330 depicted in fig. 13)
obtaining probability distributions of the attributes to be decoded at the current iteration; (¶64 and fig. 13, “decodes an occupancy symbol for the current node using an adaptive entropy decoder based on the predicted occupancy symbol distribution” at 1340 depicted in fig. 13 as part of an iteration of the “iteratively repeating 1310 through 1350 for all nodes”)
But does not explicitly teach,
determining attributes to be decoded at a current iteration;
obtaining a bitstream of the attributes to be decoded at the current iteration;
decoding, from the bitstream, with arithmetic decoding, the attributes to be decoded at the current iteration,
wherein decoding the attributes is based on the probability distributions of the attributes to be decoded at the current iteration.
However, Yang teaches additionally,
determining attributes to be decoded at a current iteration; (¶57,95, and fig. 7, “geometry information of the point cloud is first decoded” and used as additional information into an “attribute decoder to assist in decompression of the point cloud attribute” while the decoding end “successively parses a bitstream” when geometry decoding)
obtaining a bitstream of the attributes to be decoded at the current iteration; (¶88,95, and fig. 7, “attribute bitstream” within the binary bitstream are first decoded independently by point cloud decoder depicted in fig. 7) and
decoding, from the bitstream, (¶88, “decoding the attribute bitstream”) with arithmetic decoding, the attributes to be decoded (¶88, when decoding the attribute bitstream, “attribute information of the point cloud is obtained through arithmetic decoding”) at the current iteration, (¶88 and 95, point cloud decoding “successively parses a bitstream to reconstruct the prediction tree structure” of “each node”)
It would have been obvious to one with ordinary skill in the art before the effective filing date of the claimed invention to combine the cloud processing of Lodhi with the point cloud coding of Yang which uses geometry information to determine attribute information. This added teaching can simplify the coding operations, providing improved performance to point cloud coding.
Krishnan teaches additionally,
wherein decoding the attributes (¶130, “decompresser” decodes “the symbol”) is based on the probability distributions of the attributes (¶130, decompressor “updates its model with the decoded output symbol” acting identical to compressor “probability distribution (prediction)”) to be decoded at the current iteration. (¶130, decompresser “decodes the symbol” by making identical prediction using data it has already decoded and “updates its model with the decoded output symbol”)
It would have been obvious to one with ordinary skill in the art before the effective filing date of the claimed invention to combine the cloud processing of Lodhi with the point cloud coding of Yang with the probability updating of Krishnan which uses past inputs to estimate the next symbol. This allows for developing a model which can improve coding efficiency in some circumstances.
Regarding claim 10, it is the apparatus claim of method claim 1.
Lodhi teaches additionally,
An apparatus (¶66 and fig. 10, “system 1000 includes at least one processor 1010 configured to execute instructions”) comprising:
a processor; (¶66 and fig. 10, “processor 1010” depicted in fig. 10) and
a memory storing instructions operative, (¶66-68 and fig. 10, Program code to be loaded onto processor 1010 that is “stored in storage device 1040” depicted in fig. 10) when executed by the processor, (¶66, “processor 1010 configured to execute instructions loaded therein”) to cause the apparatus to:
See the mapping of claim 1 to teach the additional limitations of claim 10.
Regarding claim 11, Lodhi teaches,
A method of iteratively encoding point cloud data at an octree level, (¶63 and fig. 12, “Operations at 1230 through 1250 are repeated iteratively for every node” of a raw point cloud geometry data converted into the “octree structure”) comprising:
obtaining already-encoded attributes of a current level; (¶63 and fig. 12, “operation at 1200 receives data representing a point cloud” and “compresses the data” before being converted into a “tree representation” as depicted in fig. 12)
predicting probability distributions of attributes (¶63 and fig. 12, “occupancy symbol distribution for a current node is predicted” at 1230 depicted in fig. 12) not encoded in the current level, (¶63,54, and fig. 12, occupancy symbol distribution for a current node is predicted based on “one or more ancestor nodes of the current node” such as a parent level)
wherein the predicting (¶63 and fig. 12, “occupancy symbol distribution for a current node is predicted”) is based on the already-encoded attributes of the current level (¶63 and fig. 12, occupancy symbol distribution “feature information from one or more available neighboring nodes” at 1230, where the neighboring nodes have already been iterated through the iteratively repeated “1230 through 1240” up to the current node) and previous levels; (¶63 and fig. 12, occupancy symbol distribution based on feature information “from one or more ancestor nodes of the current node” at 1230 depicted in fig. 12)
obtaining probability distributions of the attributes to be encoded at the current iteration; (¶63 and fig. 12, “encodes an occupancy symbol for the current node” using an adaptive entropy encoder “based on the predicted occupancy symbol distribution” at 1240 depicted in fig. 12 as part of an iteration of the “1230 through 1240 repeated iteratively for every node”)
But does not explicitly teach,
determining attributes to be encoded at a current iteration;
encoding, into a bitstream, with arithmetic encoding, the attributes to be encoded at the current iteration,
wherein encoding the attributes is based on the probability distributions of the attributes to be encoded at the current iteration.
However, Yang teaches additionally,
determining attributes to be encoded at a current iteration; (¶87,94 and fig. 4, “transform color information (i.e., attribute information)” for a point in a “partitioned leaf node” corresponding to reconstructed geometry information at an “iteration”)
encoding, into a bitstream, (¶87, “arithmetic encoding is performed” to generate “a binary attribute bitstream”) with arithmetic encoding, the attributes (¶87, “arithmetic encoding is performed on quantized coefficients to generate a binary attribute bitstream”) to be encoded at the current iteration, (¶87 and 94, point cloud encoding “through successively iteration” to generate a binary stream “performed on a point”)
It would have been obvious to one with ordinary skill in the art before the effective filing date of the claimed invention to combine the cloud processing of Lodhi with the point cloud coding of Yang which uses geometry information to determine attribute information. This added teaching can simplify the coding operations, providing improved performance to point cloud coding.
Krishnan teaches additionally,
wherein encoding the attributes (¶130, “compressor” for the “next symbol”) is based on the probability distributions (¶130, “probability distribution (prediction)”) of the attributes to be encoded at the current iteration. (¶130, “compressor uses past input to estimate a probability distribution (prediction) for the next symbol”)
It would have been obvious to one with ordinary skill in the art before the effective filing date of the claimed invention to combine the cloud processing of Lodhi with the point cloud coding of Yang with the probability updating of Krishnan which uses past inputs to estimate the next symbol. This allows for developing a model which can improve coding efficiency in some circumstances.
Claim(s) 2-3,12-14 rejected under 35 U.S.C. 103 as being unpatentable over Lodhi; Muhammad Asad et al. (US 20240078715 A1) in view of YANG; Fuzheng et al. (US 20260032261 A1) in view of KRISHNAN; Madhu Peringassery et al. (US 20240080446 A1) in view of MOHANANCHETTIAR; Arunkumar et al. (US 20240163485 A1)
Regarding claim 2, Lodhi with Yang with Krishnan teaches the limitations of claim 1,
But does not explicitly teach the additional limitations of claim 2,
However, Mohananchettiar teaches additionally,
determining attributes to be decoded at the current iteration (¶46 and Fig. 6B, “decoding high-level syntax for the entropy model of the latent features” depicted in fig. 6B) comprises performing attribute selection to select the attributes to be decoded at the current iteration, (¶46 and Fig. 6B, “retrieve a deep neural network model” based on identified “PDFs being used and the extend where the selected probability distributions share their model parameters” at step 655 as depicted in fig. 6B) and
wherein performing attribute selection (¶46 and Fig. 6B, “retrieve a deep neural network model”) is based on the probability distributions of the attributes to be decoded at the current iteration. (¶46 and Fig. 6B, retrieve a deep neural network model using decoder identified PDFs being used and where the extend where the selected probability distributions share their model parameters”)
It would have been obvious to one with ordinary skill in the art before the effective filing date of the claimed invention to combine the cloud processing of Lodhi with the point cloud coding of Yang with the probability updating of Krishnan with the distribution modeling of Mohananchettiar which identifies probability distribution functions that share model parameters. This allows for support for a learning-based image or video codec that improves the compression efficiency.
Regarding claim 3, Lodhi with Yang with Krishnan with Mohananchettiar teaches the limitations of claim 2,
Mohananchettiar teaches additionally,
performing attribute deserialization (¶46 and fig. 6B, “decode the coded bitstream”) using the attributes selected to be decoded. (¶46 and fig. 6B, “generated entropy model for the latent features”, which has all the required parameters, “is applied to decode the coded bitstream”)
Regarding claim 12, Lodhi with Yang with Krishnan teaches the limitations of claim 11,
But does not explicitly teach the additional limitations of claim 12,
However, Mohananchettiar teaches additionally,
performing attribute selection to select the attributes to be encoded; (¶45 and fig. 6A, “encoder needs to decide the extent” to which the PDFs share their model parameters” at step 610 depicted in fig. 6A) and
performing attribute serialization (¶45 and fig. 6A, “generate high-level syntax” at step 615 depicted in fig. 6A) using the attributes remaining to be encoded. (¶45 and fig. 6A, generate high-level syntax using “selected PDFs and information about the sharing of their model parameters”)
It would have been obvious to one with ordinary skill in the art before the effective filing date of the claimed invention to combine the cloud processing of Lodhi with the point cloud coding of Yang with the probability updating of Krishnan with the distribution modeling of Mohananchettiar which identifies probability distribution functions that share model parameters. This allows for support for a learning-based image or video codec that improves the compression efficiency.
Regarding claim 13, Lodhi with Yang with Krishnan with Mohananchettiar teaches the limitations of claim 12,
Mohananchettiar teaches additionally,
performing attribute selection (¶45 and fig. 6A, encoder needs to “decide the extent” to which the PDFs “share their model parameters”) is based on the probability distributions of the attributes to be encoded at the current iteration. (¶45 and fig. 6A, encoder decides the extent “to which the selected probability distributions share their model parameters”)
Regarding claim 14, Lodhi with Yang with Krishnan with Mohananchettiar teaches the limitations of claim 12,
Mohananchettiar teaches additionally,
performing attribute serialization (¶45 and fig. 6A, “generate high-level syntax” at step 615 depicted in fig. 6A) generates a vector of attributes (¶45 and fig, 6A, high-level “syntax” can be updated at “the sequence level, the picture level, or sub-picture levels”) selected to be encoded at the current iteration. (¶45 and fig, 6A, generated high-level “syntax” updated at the sequence level, the picture level, or sub-picture levels as it relates to “selected PDFs and information about the sharing of their model parameters”)
Claim(s) 4 rejected under 35 U.S.C. 103 as being unpatentable over Lodhi; Muhammad Asad et al. (US 20240078715 A1) in view of YANG; Fuzheng et al. (US 20260032261 A1) in view of KRISHNAN; Madhu Peringassery et al. (US 20240080446 A1) in view of MOHANANCHETTIAR; Arunkumar et al. (US 20240163485 A1) in view of Fernandes; Felix Carlos et al. (US 20120051432 A1)
Regarding claim 4, Lodhi with Yang with Krishnan in view of Mohananchettiar teaches the limitations of claim 3,
But does not explicitly teach the additional limitations of claim 4,
However, Fernandes teaches additionally,
outputting a feedback loop of the attributes (¶50 and fig. 9, process 900 “performed a feedback loop (i.e. multiple iterations) for each input vector”) to be decoded at the current iteration. (¶50 and fig. 9, for the “input vector”)
It would have been obvious to one with ordinary skill in the art before the effective filing date of the claimed invention to combine the cloud processing of Lodhi with the point cloud coding of Yang with the probability updating of Krishnan with the distribution modeling of Mohananchettiar with the prediction of Fernandes which includes a feedback loop. This allows for repetition of the performed iterations until the highest resolution version of the frame is consistent with the measurements recovered.
Claim(s) 5-6,15-16 rejected under 35 U.S.C. 103 as being unpatentable over Lodhi; Muhammad Asad et al. (US 20240078715 A1) in view of YANG; Fuzheng et al. (US 20260032261 A1) in view of KRISHNAN; Madhu Peringassery et al. (US 20240080446 A1) in view of MOHANANCHETTIAR; Arunkumar et al. (US 20240163485 A1) in view of Mittal; Udar et al. (US 20160049156 A1)
Regarding claim 5, Lodhi with Yang with Krishnan with Mohananchettiar teaches the limitations of claim 2,
But does not explicitly teach the additional limitations of claim 5,
However, Mittal teaches additionally,
concatenating the attributes remaining to be decoded; (¶74, “previously coded elements, and their associated probabilities”)
performing feature aggregation (¶74, “associated probabilities are aggregated” of previously coded elements “are aggregated”) on the concatenated attributes remaining to be decoded, (¶74, “probability combinations of future coded elements are examined”)
wherein the feature aggregation (¶74, “calculate the conditional probabilities”) generates a score for each of the attributes remaining to be decoded; (¶74, calculating the conditional probabilities which “adjust the probabilities of the remaining quantized vector elements to be coded”) and
sorting the scores into an order for decoding the attributes, (¶75, the probabilities “may be sorted in descending order” as depicted by the “recursive process”)
wherein decoding the attributes (¶79 and 74-75, “code the non-zero positions” associated with the probabilities sorted in descending order) is further based on the order for decoding the attributes. (¶79 and 74-75, “remaining quantized vector elements to be coded” with non-zero positions sorted by “probabilities” in “descending order” which increases the chances of early termination)
Mittal discloses the aggregation of probabilities associated with coded elements. The prior art uses this aggregated probability to calculate the conditional probabilities by adjusting the remaining quantized vector elements. It would have been obvious to one with ordinary skill in the art before the effective filing date of the claimed invention to combine the cloud processing of Lodhi with the point cloud coding of Yang with the probability updating of Krishnan with the distribution modeling of Mohananchettiar with the method of Mittal that uses the statistical properties of previously coded elements. This allows for reduced per vector probability computations.
Regarding claim 6, Lodhi with Yang with Krishnan with Mohananchettiar teaches the limitations of claim 2,
But does not explicitly teach the additional limitations of claim 6,
However, Mittal teaches additionally,
generating a confidence level for each of the attributes remaining to be decoded; (¶74, “calculate the conditional probabilities” by adjusting the “probabilities of the remaining quantized vector elements to be coded) and
sorting the confidence levels into an order for decoding the attributes, (¶75, probabilities may be “sorted in descending order”)
wherein decoding the attributes (¶79 and 74-75, “code the non-zero positions” associated with the probabilities sorted in descending order) is further based on the order for decoding the attributes. (¶79 and 74-75, “remaining quantized vector elements to be coded” with non-zero positions sorted by “probabilities” in “descending order” which increases the chances of early termination)
Mittal discloses the aggregation of probabilities associated with coded elements. The prior art uses this aggregated probability to calculate the conditional probabilities by adjusting the remaining quantized vector elements. It would have been obvious to one with ordinary skill in the art before the effective filing date of the claimed invention to combine the cloud processing of Lodhi with the point cloud coding of Yang with the probability updating of Krishnan with the distribution modeling of Mohananchettiar with the method of Mittal that uses the statistical properties of previously coded elements. This allows for reduced per vector probability computations.
Regarding claim 15, dependent on claim 12, it is the encoding method of decoding claim 5, dependent on claim 2. Refer to rejection of claim 5 to teach the limitations of claim 15.
Regarding claim 16, dependent on claim 12, it is the encoding method of decoding claim 6, dependent on claim 2. Refer to rejection of claim 6 to teach the limitations of claim 16.
Claim(s) 7-8,17,19 rejected under 35 U.S.C. 103 as being unpatentable over Lodhi; Muhammad Asad et al. (US 20240078715 A1) in view of YANG; Fuzheng et al. (US 20260032261 A1) in view of KRISHNAN; Madhu Peringassery et al. (US 20240080446 A1) in view of AKHTAR; Anique et al. (US 20220012945 A1)
Regarding claim 7, Lodhi with Yang with Krishnan teaches the limitations of claim 1,
But does not explicitly teach the additional limitations of claim 7,
However, Akhtar teaches additionally,
the attributes comprise voxel occupancies. (¶35 and fig. 3A, “occupancy map 304”, depicted in fig. 3A, corresponding to “a prediction probability of that voxel being occupied”)
It would have been obvious to one with ordinary skill in the art before the effective filing date of the claimed invention to combine the cloud processing of Lodhi with the point cloud coding of Yang with the probability updating of Krishnan with the point cloud processing of Akhtar that indicates a prediction probability for a voxel. This allows for greater levels of compression by allowing increased level of detail for point cloud data after quantization.
Regarding claim 8, Lodhi with Yang with Krishnan with Akhtar teaches the limitations of claim 7,
Akhtar teaches additionally,
determining attributes to be decoded at the current iteration (¶35 and fig. 3A, “point cloud processing” depicted in fig. 3A) comprises performing voxel selection (¶35, “choose top k voxels or have a threshold defining a voxel being occupied”) to select voxel occupancies to be decoded at the current iteration, (¶35, “choose top k voxels or have a threshold defining a voxel being occupied if the prediction value is greater than that threshold” based on an occupancy map 304 depicted in fig. 3A) and
wherein performing voxel selection (¶35 and fig. 3A, choose “top K voxels” or “voxel being occupied if the prediction value is greater than that threshold”) is based on the probability distributions of the attributes to be decoded at the current iteration. (¶35 and fig. 3A, choose that “voxel being occupied”, corresponding to the occupancy map 304, if the prediction probability value “of that voxel” is greater than a threshold)
It would have been obvious to one with ordinary skill in the art before the effective filing date of the claimed invention to combine the cloud processing of Lodhi with the point cloud coding of Yang with the probability updating of Krishnan with the point cloud processing of Akhtar that indicates a prediction probability for a voxel. This allows for greater levels of compression by allowing increased level of detail for point cloud data after quantization.
Regarding claim 17, dependent on claim 11, it is the encoding claim of decoding claim 7, dependent on claim 1. Refer to rejection of claim 7 to teach the limitations of claim 17.
Regarding claim 19, Lodhi with Yang with Krishnan with Akhtar teaches the limitations of claim 17,
Akhtar teaches additionally,
performing voxel selection (¶35 and fig. 3A, choose “top K voxels” or “voxel being occupied if the prediction value is greater than that threshold”) is based on probability distributions of the voxel occupancies to be encoded at the current iteration. (¶35 and fig. 3A, choose that “voxel being occupied”, corresponding to the occupancy map 304, if the prediction probability value “of that voxel” is greater than a threshold)
It would have been obvious to one with ordinary skill in the art before the effective filing date of the claimed invention to combine the cloud processing of Lodhi with the point cloud coding of Yang with the probability updating of Krishnan with the point cloud processing of Akhtar that indicates a prediction probability for a voxel. This allows for greater levels of compression by allowing increased level of detail for point cloud data after quantization.
Claim(s) 9,18,20 rejected under 35 U.S.C. 103 as being unpatentable over Lodhi; Muhammad Asad et al. (US 20240078715 A1) in view of YANG; Fuzheng et al. (US 20260032261 A1) in view of KRISHNAN; Madhu Peringassery et al. (US 20240080446 A1) in view of AKHTAR; Anique et al. (US 20220012945 A1) in view of Pham Van; Luong et al. (US 20230018907 A1)
Regarding claim 9, Lodhi with Yang with Krishnan with Akhtar teaches the limitations of claim 8,
But does not explicitly teach the additional limitations of claim 9,
However, Pham Van teaches additionally,
performing voxel deserialization (¶101 and fig. 4, “Octree synthesis unit 306 may synthesize an octree” depicted in fig. 4) using the voxel occupancies selected to be decoded. (¶101, syntax elements “parsed from the geometry bitstream” decoded by geometry arithmetic decoding unit 302 signaling the “occupancy of each of the eight children node”)
It would have been obvious to one with ordinary skill in the art before the effective filing date of the claimed invention to combine the cloud processing of Lodhi with the point cloud coding of Yang with the probability updating of Krishnan with the point cloud processing of Akhtar with the occupancy data of Pham Van that parses the signaled occupancy for nodes. This allows for techniques that can reduce the processing performed and the size of the bitstream.
Regarding claim 18, Lodhi with Yang with Krishnan with Akhtar teaches the limitations of claim 17,
Akhtar teaches additionally,
performing voxel selection to select voxel occupancies to be encoded; (¶35, “choose top k voxels or have a threshold defining a voxel being occupied if the prediction value is greater than that threshold” based on an occupancy map 304 depicted in fig. 3A)
But does not teach the additional limitation of claim 18.
However, Pham Van teaches additionally,
performing voxel serialization using voxel occupancies (¶75 and fig. 2, octree analysis unit 210, depicted in fig. 2, may “store data representing occupied voxels (i.e., voxels occupied by points of the point cloud)”) remaining to be encoded. (¶75,77, and fig. 2, for multiple points of the point cloud stored in memory 228 which allows arithmetic encoding unit 214 to “encode occupancy data”)
It would have been obvious to one with ordinary skill in the art before the effective filing date of the claimed invention to combine the cloud processing of Lodhi with the point cloud coding of Yang with the probability updating of Krishnan with the point cloud processing of Akhtar with the occupancy data of Pham Van that is encoded using arithmetic encoding. This allows for techniques that can reduce the processing performed and the size of the bitstream.
Regarding claim 20, Lodhi with Yang with Krishnan with Akhtar teaches the limitations of claim 17,
But does not explicitly teach the additional limitations of claim 20,
However, Pham teaches additionally,
performing voxel serialization generates a vector (¶75 and fig. 2, octree analysis unit 210, depicted in fig. 2, may “store data representing occupied voxels (i.e., voxels occupied by points of the point cloud)”) of the voxel occupancies (¶75, “occupied voxels”) selected to be encoded at the current iteration. (¶75, “voxels occupied by points of the point cloud”)
It would have been obvious to one with ordinary skill in the art before the effective filing date of the claimed invention to combine the cloud processing of Lodhi with the point cloud coding of Yang with the probability updating of Krishnan with the point cloud processing of Akhtar with the occupancy data of Pham Van that is encoded using arithmetic encoding. This allows for techniques that can reduce the processing performed and the size of the bitstream.
Conclusion
THIS ACTION IS MADE FINAL. 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 JIMMY S LEE whose telephone number is (571)270-7322. The examiner can normally be reached Monday thru Friday 10AM-8PM EST.
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, Joseph G. Ustaris can be reached at (571) 272-7383. 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.
/JOSEPH G USTARIS/Supervisory Patent Examiner, Art Unit 2483
/JIMMY S LEE/Examiner, Art Unit 2483
Prosecution Insights