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 .
Priority
Acknowledgement is made of Applicant’s claim of the present application claiming priority and benefit under 35 U.S.C. 119(a-d) to Korean Patent Application No. KR10-2024-0111773 filed 08/21/2024 and Korean Patent Application No. KR10-2023-0125068 filed 09/19/2023. It appears that the priority document for Korean Patent Application No. KR10-2023-0125068 that has been retrieved, is incorrect (a presentation rather than a Korean Patent document).
Information Disclosure Statement
The information disclosure statement (“IDS”) filed on 09/19/2024 has been reviewed and the listed references were noted.
Drawings
The 25-page drawings have been considered and placed in the file.
Status of Claims
Claims 1-12 are pending.
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.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1, 2, 4-7, and 10-12 are rejected under 35 U.S.C. 103 as being unpatentable over Tang et al. (“Deep Implicit Volume Compression”), in view of Mao et al. (WO 2025011009 A1 w/ EFD of 07/11/2023).
Regarding claim 1, Tang teaches, “A method for encoding a TSDF volume, the method comprising: lossily encoding magnitude information of a Truncated Signed Distance Field (TSDF) volume (Tang, Figure 3 Caption discloses; “Therefore, the average reconstructed error due to lossy magnitude compression is bounded by the size of a voxel (5mm).” Examiner interprets this disclosure to imply that Tang lossily encodes the magnitude information.) “and losslessly encoding sign information of the TSDF volume (Tang, Pages 1-2 discloses; “Furthermore, we propose using the conditional distribution of the signs given 1 the encoded TSDF block to compress the signs losslessly, leading to significant gains in bitrates.”) “(Mao, Page 34 discloses; “The obtained feature map y is input into the hyper-network to obtain the information z (i.e., the first information mentioned above). The quantized z is entropy-encoded by me-tANS and written into the bitstream.” Examiner interprets that this disclosure teaches that the hyperprior model selects information and entropy-encodes this information based on the hyperprior model.) Mao also teaches, “based on a hyperprior model” and “based on the hyperprior model” (Mao, Page 34 discloses; “The obtained feature map y is input into the hyper-network to obtain the information z (i.e., the first information mentioned above). The quantized z is -by me-tANS and written into the bitstream.” Mao teaches using a hyperprior model for encoding, thus it would be obvious to encode the sign and magnitude information of Tang based on the hyperprior model of Mao.)
Tang and Mao are considered to be analogous to the claimed invention because they are in the same field of encoding and decoding image/3D data. Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to have modified Tang to incorporate the teachings of Mao in order to use a hyperprior model to encode 3D volume information. One of ordinary skill in the art would have been motivated to combine the previously described apparatus of Tang with the teachings of Mao to include information about spatial dependencies and variances in the latent space. Accordingly, it would have been obvious to combine Tang and Mao to obtain the method of claim 1.
Regarding claim 2, the combination of Tang and Mao teaches, “The method of claim 1, wherein the lossy encoding comprises: converting the TSDF volume into a latent vector” (Tang, Page 12 “Lossy transform coding” discloses; “
PNG
media_image1.png
83
325
media_image1.png
Greyscale
”) “and converting the latent vector into a hyperprior vector for the hyperprior model;” (Tang, Page 2, “Learnable compression strategies” discloses; “This was extended by placing a hierarchical hyperprior on the latent representations to significantly improve the image compression performance [3].” Examiner interprets this as “placing a hierarchical hyperprior on the latent representation” would create a hyperprior vector.) “and decoding the hyperprior vector to generate the selection information and probability distribution information of the latent vector.” (Mao, Page 34 discloses; “Inputting the quantized hyper-prior information z` into the hyper-scale decoder network can obtain the Gaussian distribution parameter information σ (that is, the first Gaussian distribution parameter information mentioned above).”) The proposed combination as well as the motivation for combining the Tang and Mao references presented in the rejection of claim 1, apply to claim 2 and are incorporated herein by reference. Thus, the method recited in claim 2 is met by Tang and Mao.
Regarding claim 4, the combination of Tang and Mao teaches, “The method of claim 2, wherein the selected elements of the latent vector are entropy-encoded based on probability distribution information of the selected elements, among the probability distribution information of the latent vector.” (Tang, Pages 3-4 disclose; “The compression pipeline is illustrated in Figure 2. Given a block x to be transmitted, the sender first computes the lossily quantized latent representation z= E(x;θe) using the learned encoder E with parameters θe. Next, the sender uses z to compute the conditional probability distribution over the TSDF signs as ps|z (s|z;θs), where s is the ground truth sign configuration of the block, and θs are the learn able parameters of the distribution. The sender then uses an entropy coder to compute the bitstreams zbits and sbits by losslessly coding the latent code z and signs s using the distributions pz (z; φ) and ps|z (s|z; θs) respectively.” Examiner interprets that it would be obvious to combine the encoding method of Tang with the selected information of Mao.) The proposed combination as well as the motivation for combining the Tang and Mao references presented in the rejection of claim 1, apply to claim 4 and are incorporated herein by reference. Thus, the method recited in claim 4 is met by Tang and Mao.
Regarding claim 5, the combination of Tang and Mao teaches, “The method of claim 2, wherein the lossy encoding comprises generating sign probability information of the TSDF volume based on the selection information.” (Tang, Page 4, “Rate of losslessly compressed signs” discloses; “Since s contains only discrete values {−1, +1}, it can be compressed losslessly using entropy coding. As mentioned above, we use the conditional probability distribution ps|z (s|z) instead of the prior distribution ps(s).” Tang discloses that s is representative of signs. Also, it would be obvious to combine the probability of Tang with the selection information of Mao to have the sign probability based on the selection information.) The proposed combination as well as the motivation for combining the Tang and Mao references presented in the rejection of claim 1, apply to claim 5 and are incorporated herein by reference. Thus, the method recited in claim 5 is met by Tang and Mao.
Regarding claim 6, the combination of Tang and Mao teaches, “The method of claim 5, wherein the lossless encoding comprises entropy-encoding the sign information of the TSDF volume based on the sign probability information to generate a sign bitstream.” (Tang, Figure 2 Caption discloses; “Then ˆ z and s are entropy coded and transmitted to the receiver (Aenc and Adec blocks) using a prior learned distribution pz (z) and the conditional distribution psz (s|z) as estimated by the decoder, respectively” and Tang, Page 3, Para. 5 discloses; “The sender then uses an entropy coder to compute the bitstreams zbits and sbits”).
Regarding claim 7, the combination of Tang and Mao teaches, “A method for decoding a TSDF volume, the method comprising: lossily decoding magnitude information of a TSDF volume based on a hyperprior model;” (Tang, Figure 3 Caption discloses; “Therefore, the average reconstructed error due to lossy magnitude compression is bounded by the size of a voxel (5mm).” Examiner interprets this disclosure to imply that Tang lossily decodes magnitude information. Mao, Page 34 discloses; “Inputting the quantized hyper-prior information z` into the hyper-scale decoder network can obtain the Gaussian distribution parameter information σ (that is, the first Gaussian distribution parameter information mentioned above).” Mao teaches using a hyperprior model for decoding, thus it would be obvious to decode magnitude information of Tang based on the hyperprior model of Mao.) “and losslessly decoding sign information of the TSDF volume based on the hyperprior model” (Tang, Page 4, Para. 2 discloses; “As we transmit the signs losslessly, it is guaranteed that the mesh extracted from the decoded TSDF ˆx will have the same topology as the mesh extracted from the uncompressed TSDF x.” Mao, Page 34 discloses; “Inputting the quantized hyper-prior information z` into the hyper-scale decoder network can obtain the Gaussian distribution parameter information σ (that is, the first Gaussian distribution parameter information mentioned above).” Mao teaches using a hyperprior model for decoding, thus it would be obvious to decode sign information of Tang based on the hyperprior model of Mao.) “wherein the lossy decoding comprises: deriving selected elements of a latent vector of the TSDF volume from a bitstream by entropy-decoding the bitstream” (Tang, Page 4, Para. 1 discloses; “which first recovers z using entropy decoding with the shared prior pz.”) “based on probability distribution information for the latent vector, the probability distribution information being obtained through the hyperprior model;” (Tang, Page 12 “Lossy transform coding” discloses; “In lossy transform coding, a transformation is used to transform the original data x into a latent representation z=E(x;θe) and another is used to approximately recover the original data x=D(z;θd) from the lossy latent representation ˆ z.” Examiner interprets z to be the latent vector.) “and generating sign probability information and magnitude information of voxels constituting the TSDF volume based on selection information for the latent vector obtained through the hyperprior model and the selected elements of the latent vector.” (Tang, Page 4, “Rate of losslessly compressed signs” discloses; “Since s contains only discrete values {−1, +1}, it can be compressed losslessly using entropy coding. As mentioned above, we use the conditional probability distribution ps|z (s|z) instead of the prior distribution ps(s).” Tang discloses that s is representative of signs. Also, it would be obvious to combine the probability of Tang with the selection information of Mao to have the sign probability based on the selection information. Tang, Page 4 “Distortion” discloses; “We minimize the reconstruction error between the ground truth and the predicted TSDF values.” Examiner interprets this to be magnitude information.)
Tang and Mao are considered to be analogous to the claimed invention because they are in the same field of encoding and decoding image/3D data. Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to have modified Tang to incorporate the teachings of Mao in order to use a hyperprior model to decode 3D volume information. One of ordinary skill in the art would have been motivated to combine the previously described apparatus of Tang with the teachings of Mao to include information about spatial dependencies and variances in the latent space. Accordingly, it would have been obvious to combine Tang and Mao to obtain the method of claim 7.
Regarding claim 10, the combination of Tang and Mao teaches, “The method of claim 7, wherein the lossless decoding comprises obtaining the sign information of the TSDF volume by entropy-decoding the bitstream” (Tang, Page 4, Para. 1 discloses; “The receiver then re-computes ps|z in order to recover the losslessly coded ground truth signs s.”) “based on the sign probability information.” (Tang, Page 4, “Rate of losslessly compressed signs” discloses; “
PNG
media_image2.png
135
327
media_image2.png
Greyscale
”)
Regarding claim 11, the combination of Tang and Mao teaches, “The method of claim 7, further comprising obtaining the TSDF volume by multiplying the magnitude information and the sign information of the TSDF volume.” (Tang, Page 4, Para. 1 discloses; “
PNG
media_image3.png
101
327
media_image3.png
Greyscale
”)
Regarding claim 12, the combination of Tang and Mao teaches, “An apparatus for encoding a TSDF volume, the apparatus comprising: a memory that stores data and one or more instructions;” (Mao, Page 10 discloses; “Optionally, the device may further include at least one memory, and the at least one memory is used to store the program code or instruction.”) “and one or more processors for executing the one or more instructions stored in the memory, wherein, by executing the one or more instructions” (Mao, Pages 9-10 discloses; “In the tenth aspect, an embodiment of the present application also provides a coding device, which includes: at least one processor, when the at least one processor executes program code or instructions, it implements the method described in the above first aspect or any possible implementation method thereof.”) “the one or more processors are configured to: lossily encode magnitude information of a Truncated Signed Distance Field (TSDF) volume based on a hyperprior model” (Tang, Figure 3 Caption discloses; “Therefore, the average reconstructed error due to lossy magnitude compression is bounded by the size of a voxel (5mm).” Examiner interprets this disclosure to imply that Tang lossily encodes the magnitude information.) “wherein some elements of a latent vector for the TSDF volume are selected and entropy-encoded based on selection information obtained through the hyperprior model” (Mao, Page 34 discloses; “The obtained feature map y is input into the hyper-network to obtain the information z (i.e., the first information mentioned above). The quantized z is -by me-tANS and written into the bitstream.” Mao teaches using a hyperprior model for encoding, thus it would be obvious to encode the sign and magnitude information of Tang based on the hyperprior model of Mao.) “and losslessly encode sign information of the TSDF volume based on the hyperprior model,” (Tang, Pages 1-2 discloses; “Furthermore, we propose using the conditional distribution of the signs given 1 the encoded TSDF block to compress the signs losslessly, leading to significant gains in bitrates.”) “herein the sign information of the TSDF volume is entropy-encoded based on probability distribution information of the latent vector obtained through the hyperprior model.” (Tang, Pages 3-4 disclose; “The compression pipeline is illustrated in Figure 2. Given a block x to be transmitted, the sender first computes the lossily quantized latent representation z= E(x;θe) using the learned encoder E with parameters θe. Next, the sender uses z to compute the conditional probability distribution over the TSDF signs as ps|z (s|z;θs), where s is the ground truth sign configuration of the block, and θs are the learn able parameters of the distribution. The sender then uses an entropy coder to compute the bitstreams zbits and sbits by losslessly coding the latent code z and signs s using the distributions pz (z; φ) and ps|z (s|z; θs) respectively.” Examiner interprets that it would be obvious to combine the encoding method of Tang with the selected information of Mao.)
Tang and Mao are considered to be analogous to the claimed invention because they are in the same field of encoding and decoding image/3D data. Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to have modified Tang to incorporate the teachings of Mao in order to use a hyperprior model to encode 3D volume information based on a probability distribution. One of ordinary skill in the art would have been motivated to combine the previously described apparatus of Tang with the teachings of Mao to include information about spatial dependencies and variances in the latent space. Accordingly, it would have been obvious to combine Tang and Mao to obtain the method of claim 12.
Claims 3, 8, and 9 are rejected under 35 U.S.C. 103 as being unpatentable over Tang et al. (“Deep Implicit Volume Compression”), in view of Mao et al. (WO 2025011009 A1 w/ EFD of 07/11/2023), in further view of Ballé et al. (“Variational Image Compression with a Scale Hyperprior”).
Regarding claim 3, the combination of Tang and Mao teaches, “The method of claim 2, wherein the lossy encoding comprises entropy-encoding the hyperprior vector (Mao, Page 34 discloses; “The obtained feature map y is input into the hyper-network to obtain the information z (i.e., the first information mentioned above). The quantized z is entropy-encoded by me-tANS and written into the bitstream.”) The combination of Tang and Mao does not explicitly teach, “based on a distribution according to a factorized prior model”. Since the combination of Tang and Mao does not explicitly disclose this limitation, Examiner relies on the teachings of Ballé in an analogous field of endeavor. Specifically, Ballé teaches, “based on a distribution according to a factorized prior model” (Ballé, Page 2, Para. 3 discloses; “Specifically, we extend the model presented in Ballé et al. (2017), which has a fully factorized prior, with a hyperprior that captures the fact that spatially neighboring elements of the latent representation tend to vary together in their scales.” Examiner interprets that it would be obvious to combine the factorized prior model of Ballé with the encoding of Mao to have the encoding based on the distribution from the factorized prior model.)
Tang, Mao, and Ballé are considered to be analogous to the claimed invention because they are in the same field of encoding and decoding 3D/image data. Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to have modified the combination of Tang and Mao to incorporate the teachings of Ballé in order to include a factorized prior model to base the encoding on. One of ordinary skill in the art would have been motivated to combine the previously described apparatus of Tang and Mao with the teachings of Ballé to, (Ballé Page 2, “captures the fact that spatially neighboring elements of the latent representation tend to vary together in their scale”). Accordingly, it would have been obvious to combine Tang, Mao, and Ballé to obtain the method of claim 3.
Regarding claim 8, the combination of Tang, Mao, and Ballé teaches, “The method of claim 7, wherein the lossy decoding comprises: entropy-decoding the bitstream to generate a hyperprior vector for the hyperprior model; and decoding the hyperprior vector to generate the selection information and the probability distribution information.” (Ballé, Page 6, Para. 1 discloses; “The decoder first recovers ˆz from the compressed signal. It then uses hs to obtain σ, which provides it with the correct probability estimates to successfully recover y as well.” Examiner interprets z to be a hyperprior vector and sigma to be selection and probability information.) The proposed combination as well as the motivation for combining the Tang, Mao, and Ballé references presented in the rejection of claim 3, apply to claim 8 and are incorporated herein by reference. Thus, the method recited in claim 8 is met by Tang, Mao, and Ballé.
Regarding claim 9, the combination of Tang, Mao, and Ballé teaches, “The method of claim 7, wherein the generating of the sign probability information and the magnitude information comprises performing an operation on the selection information and the selected elements of the latent vector to reconstruct the latent vector to an original dimension.” (Figure 4 of Ballé discloses performing an “operation” (hs and gs in the image) on information to reconstruct a latent vector to an original dimension. It would be obvious to use the selection information of Mao combined with the operation of Ballé. Mao, Page 34 discloses; “The obtained feature map y is input into the hyper-network to obtain the information z (i.e., the first information mentioned above). The quantized z is entropy-encoded by me-tANS and written into the bitstream.” Examiner interprets that this disclosure teaches that the hyperprior model selects information and entropy-encodes this information based on the hyperprior model.) The proposed combination as well as the motivation for combining the Tang, Mao, and Ballé references presented in the rejection of claim 3, apply to claim 9 and are incorporated herein by reference. Thus, the method recited in claim 9 is met by Tang, Mao, and Ballé.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JUSTIN M. OAKES whose telephone number is (571)272-9379. The examiner can normally be reached 7:30am-5pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Amandeep Saini can be reached at (571) 272-3382. 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.
/JUSTIN M OAKES/Examiner, Art Unit 2662
/Siamak Harandi/Primary Examiner, Art Unit 2662
Prosecution Insights