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 .
Priority
Acknowledgment is made of applicant’s claim for foreign priority based on an application filed in Europe on 17 September 2021. It is noted, however, that applicant has not filed a certified copy of the EP21197459.7 application as required by 37 CFR 1.55.
Response to Amendments
The amendment filed 07 July 2025 is entered.
Claims 1, 4-6, 8-13, 15, and 18-20 are amended.
Claims 2-3, 7, 16-17 are cancelled.
Claims 1, 4-6, 8-15, and 18-20 are pending.
Claim Objections
Claim 15 is objected to because of the following informalities:
Claim 15, line 9, the word “decomposing” should be “decompose.”
Appropriate correction is required.
Claim Rejections - 35 USC § 101
35 U.S.C. § 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1, 4-6, 8-15, and 18-20 are rejected under 35 U.S.C. § 101 because the claimed invention is directed to an abstract idea without significantly more.
Regarding Claim 1:
Step 1 Analysis: Is the claim to a process, machine, manufacture or composition of matter?
Yes, the claim is to a process.
Step 2A Prong One Analysis: Does the claim recite an abstract idea, law of nature, or natural phenomenon?
Yes, the limitation “obtaining three-dimensional (3D) data, the 3D data comprising range data, angle data, and doppler data” is the abstract idea of a mental process that can practically be performed in the human mind.
Yes, the limitation “decomposing the 3D data into three sets of two-dimensional (2D) data, a first set of 2D data comprising range data and angle data, a second set of 2D data comprising range data and doppler data and a third set of 2D data comprising angle data and doppler data” is the abstract idea of a mathematical concept.
Yes, the limitation “processing a deep-learning algorithm on the 3D data to obtain processed 3D data by processing the first set of 2D data, the second set of 2D data and the third set of 2D data individually and solely by utilizing 2D convolutions” is the abstract idea of a mathematical concept.
Yes, the limitation “obtaining processed two-dimensional (2D) data from the processed 3D data, the processed 2D data comprising range data and angle data” is the abstract idea of a mental process that can practically be performed in the human mind.
Yes, the limitation “wherein processing the deep learning algorithm on the 3D data further comprises processing a position encoding algorithm on the second set of 2D data and the third set of 2D data” is the abstract idea of a mathematical concept.
Step 2A Prong Two Analysis: Does the claim recite additional elements that integrate the judicial exception into a practical application?
No.
Step 2B Analysis: Does the claim recite additional elements that amount to significantly more than the judicial exception?
No.
Regarding Claim 15:
Step 1 Analysis: Is the claim to a process, machine, manufacture or composition of matter?
Yes, the claim is to a process.
Step 2A Prong One Analysis: Does the claim recite an abstract idea, law of nature, or natural phenomenon?
Yes, the limitation “obtain 3D data, the 3D data comprising range data, angle data, and doppler data” is the abstract idea of a mental process that can practically be performed in the human mind.
Yes, the limitation “decomposing the 3D data into three sets of two-dimensional (2D) data, a first set of 2D data comprising range data and angle data, a second set of 2D data comprising range data and doppler data and a third set of 2D data comprising angle data and doppler data” is the abstract idea of a mathematical concept.
Yes, the limitation “process a deep-learning algorithm on the 3D data to obtain processed 3D data by processing the first set of 2D data, the second set of 2D data and the third set of 2D data individually and solely by utilizing 2D convolutions” is the abstract idea of a mathematical concept.
Yes, the limitation “obtain processed 2D data from the processed 3D data, the processed 2D data comprising range data and angle data” is the abstract idea of a mental process that can practically be performed in the human mind.
Yes, the limitation “wherein the computer system processes the deep learning algorithm on the 3D data further by processing a position encoding algorithm on the second set of 2D data and the third set of 2D data” is the abstract idea of a mathematical concept.
Step 2A Prong Two Analysis: Does the claim recite additional elements that integrate the judicial exception into a practical application?
No, the limitation “a radar device; a processing device; and a non-transitory computer-readable medium storing one or more programs, the one or more programs comprising instructions, which when executed by the processing device, cause the computer system to:” are additional elements that amounts to adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer.
The additional elements, taken alone or in combinations, fail to integrate the judicial exception into a practical application.
Step 2B Analysis: Does the claim recite additional elements that amount to significantly more than the judicial exception?
No, the limitation “a radar device; a processing device; and a non-transitory computer-readable medium storing one or more programs, the one or more programs comprising instructions, which when executed by the processing device, cause the computer system to:” are additional elements that amounts to adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer.
The additional elements, taken alone or in combinations, fail to amount to significantly more than the judicial exception.
Regarding Claims 4-6, 8-14, and 18-20, the claims merely expand on the abstract ideas in the independent claims. Therefore, Claims 4-6, 8-14, and 18-20 are also rejected.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C.
102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the
statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a
new ground of rejection if the prior art relied upon, and the rationale supporting the rejection,
would be the same under either status.
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 1, 4-5, 8, 10-13, 15, and 18-19 are rejected under 35 U.S.C. 103 as being unpatentable over Major (B. Major et al., “Vehicle Detection With Automotive Radar Using Deep Learning on Range-Azimuth-Doppler Tensors,” 2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW), Seoul, Korea (South), 2019) in view of Lin (T. -Y. Lin et al., “Feature Pyramid Networks for Object Detection,” 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA, 2017).
Regarding Claim 1, Major teaches:
A computer-implemented method comprising ([Section 4]: “radar perception system”):
obtaining three-dimensional (3D) data, the 3D data comprising range data, angle data, and doppler data ([Section 1]: “The radar data is a 3D tensor, with the first two dimensions making up range-azimuth (polar) space, and the third Doppler dimension which contains velocity information.”; [Section 4]);
decomposing the 3D data into three sets of two-dimensional (2D) data, a first set of 2D data comprising range data and angle data, a second set of 2D data comprising range data and doppler data and a third set of 2D data comprising angle data and doppler data ([Section 4.1.2]: “three 2D inputs: range-azimuth, azimuth-Doppler and range-Doppler.”; Figure 3);
processing a deep-learning algorithm on the 3D data to obtain processed 3D data by processing the first set of 2D data, the second set of 2D data and the third set of 2D data individually … ([Section 2]: “the resulting tensor is the input to the machine learning model”; [Section 4.1.2]: “three 2D inputs: range-azimuth, azimuth-Doppler and range-Doppler.”; Figure 3); and
obtaining processed two-dimensional (2D) data from the processed 3D data, the processed 2D data comprising range data and angle data ([Section 4.1.1]; [Section 4.1.2]: “After these convolutions, we perform max-pooling over the Doppler dimension and continue with the up-sampling layers of the range-azimuth model, as described in the RA model section.”);
wherein processing the deep learning algorithm on the 3D data further comprises processing a position encoding algorithm on the second set of 2D data and the third set of 2D data ([Section 4.1]: “In practice, this means stacking two additional channels to the input which contain the pixel coordinates to enable the convolutions to be conditioned on location.”; [Section 4.1.2]: “three 2D inputs: range-azimuth, azimuth-Doppler and range-Doppler”).
Major does not explicitly teach: processing the first, second, and third sets of 2D data solely by utilizing 2D convolutions.
However, Major’s architecture processes the 2D data using convolutions ([Section 4.1.1]; Figure 3), and it is well-know, if not inherent, to use 2D convolutions to process 2D data as claimed. For example, see Lin ([Section 4.1]: “3×3 conv and two sibling 1×1 convs”; [Section 5.1.1]), which motivated the architecture of Major and uses 2D convolutions to process 2D data.
If not inherent, it would have been obvious to one of ordinary skill in the art to modify Major and processing the first, second, and third sets of 2D data solely by utilizing 2D convolutions, as taught by Lin. Using 2D convolutions to process 2D data is considered ordinary and well-known in the art, and it is a matter of applying a known technique to a known device ready for improvement to yield predictable results.
Regarding Claim 15, Major teaches:
A computer system comprising: a radar device; a processing device; and a non-transitory computer-readable medium storing one or more programs, the one or more programs comprising instructions ([Section 4]: “radar perception system”), which when executed by the processing device, cause the computer system to:
obtain 3D data, the 3D data comprising range data, angle data, and doppler data ([Section 1]: “The radar data is a 3D tensor, with the first two dimensions making up range-azimuth (polar) space, and the third Doppler dimension which contains velocity information.”; [Section 4]);
decomposing the 3D data into three sets of two-dimensional (2D) data, a first set of 2D data comprising range data and angle data, a second set of 2D data comprising range data and doppler data and a third set of 2D data comprising angle data and doppler data ([Section 4.1.2]: “three 2D inputs: range-azimuth, azimuth-Doppler and range-Doppler.”; Figure 3);
process a deep-learning algorithm on the 3D data to obtain processed 3D data by processing the first set of 2D data, the second set of 2D data and the third set of 2D data individually … ([Section 2]: “the resulting tensor is the input to the machine learning model”; [Section 4.1.2]: “three 2D inputs: range-azimuth, azimuth-Doppler and range-Doppler.”; Figure 3); and
obtain processed 2D data from the processed 3D data, the processed 2D data comprising range data and angle data ([Section 4.1.1]; [Section 4.1.2]: “After these convolutions, we perform max-pooling over the Doppler dimension and continue with the up-sampling layers of the range-azimuth model, as described in the RA model section.”);
wherein the computer system processes the deep learning algorithm on the 3D data further by processing a position encoding algorithm on the second set of 2D data and the third set of 2D data ([Section 4.1]: “In practice, this means stacking two additional channels to the input which contain the pixel coordinates to enable the convolutions to be conditioned on location.”; [Section 4.1.2]: “three 2D inputs: range-azimuth, azimuth-Doppler and range-Doppler”).
Major does not explicitly teach: processing the first, second, and third sets of 2D data solely by utilizing 2D convolutions.
However, Major’s architecture processes the 2D data using convolutions ([Section 4.1.1]; Figure 3), and it is well-know, if not inherent, to use 2D convolutions to process 2D data as claimed. For example, see Lin ([Section 4.1]: “3×3 conv and two sibling 1×1 convs”; [Section 5.1.1]), which motivated the architecture of Major and uses 2D convolutions to process 2D data.
If not inherent, it would have been obvious to one of ordinary skill in the art to modify Major and processing the first, second, and third sets of 2D data solely by utilizing 2D convolutions, as taught by Lin. Using 2D convolutions to process 2D data is considered ordinary and well-known in the art, and it is a matter of applying a known technique to a known device ready for improvement to yield predictable results.
Regarding Claims 4 and 18, Major discloses: wherein processing of the first set of 2D data comprises processing a compression algorithm ([Section 4.1.1]: “multiple downsampling (i.e. strided convolutional) layers”).
Regarding Claim 5 and 19, Major discloses: wherein processing at least one of the first set of 2D data, the second set of 2D data, or the third set of 2D data comprises processing a convolution algorithm ([Section 4.1.1]: “multiple consecutive convolutional layers”).
Regarding Claim 8, Major discloses: wherein processing the deep-learning algorithm on the 3D data further comprises aligning the first set of 2D data, the second set of 2D data, and the third set of 2D data in the first set of 2D data ([Section 4.1.2]: “We then concatenate these in the channel dimension and apply 3D convolutional layers.”).
Regarding Claim 10, Major discloses: wherein processing the deep-learning algorithm on the 3D data comprises processing a convolution algorithm ([Section 4.1.1]; [Section 4.1.2]: “convolutional layers”).
Regarding Claim 11, Major discloses: wherein the convolution algorithm processes the angle data and the doppler data ([Section 4.1.2]: “azimuth-Doppler”; Figure 3).
Regarding Claim 12, Major discloses: wherein the convolution algorithm processes the range data and the angle data ([Section 4.1.1]; [Section 4.1.2]: “range-azimuth”; Figure 3).
Regarding Claim 13, Major discloses: wherein processing the deep-learning algorithm on the 3D data further comprises processing an up-sample algorithm ([Section 4.1.1]: “up-sampling”).
Claim 6 rejected under 35 U.S.C. 103 as being unpatentable over Major (B. Major et al., “Vehicle Detection With Automotive Radar Using Deep Learning on Range-Azimuth-Doppler Tensors,” 2019) and Lin (T. -Y. Lin et al., “Feature Pyramid Networks for Object Detection,” 2017), as applied to Claim 1 above, and further in view of Unnikrishknan (US 2020/0217950).
Regarding Claim 6, Major does not explicitly teach – but Unnikrishknan teaches: wherein processing of the first set of 2D data comprises processing a dropout algorithm (Unnikrishknan [0086]: “Dropout layer 951 is so named for its regularizing function to randomly remove variables of the CNN during training.”). It would have been obvious to modify Major and use a dropout algorithm, as taught by Unnikrishknan. Dropout algorithms are well-known in the art and are beneficial for reducing overfitting in neural networks.
Claim 9 and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Major (B. Major et al., “Vehicle Detection With Automotive Radar Using Deep Learning on Range-Azimuth-Doppler Tensors,” 2019) and Lin (T. -Y. Lin et al., “Feature Pyramid Networks for Object Detection,” 2017), as applied to Claims 1 and 15 above, and further in view of Tan (H. Tan et al., “LXMERT: Learning Cross-Modality Encoder Representations from Transformers,” 2019).
Regarding Claims 9 and 20, Major teaches: wherein processing the deep-learning algorithm on the 3D data further comprises aligning the first set of 2D data, the second set of 2D data, and the third set of 2D data in the first set of 2D data… ([Section 4.1]: “In practice, this means stacking two additional channels to the input which contain the pixel coordinates to enable the convolutions to be conditioned on location.”).
Major does not explicitly teach – but Tan teaches: … by applying a cross-attention algorithm that uses the first set of 2D data to attend to the second set of 2D data and the third set of 2D data (Tan [Section 2.2]: “Cross-Modality Encoder”). Modifying Major to use a cross-attention algorithm that uses the first set of 2D data to attend to the second set of 2D data and the third set of 2D data would be obvious to try. There is a need to improve feature extraction and object detection in deep-learning-based vehicle detection systems (Major [Abstract]; [Section 4]). Cross-attention algorithms are well-known in the art and are beneficial for improving feature extraction and object detection (Tan [Sections 2.2, 4.3]). Major performs feature extraction over the first, second, and third sets of 2D data, which presents a finite number of predictable solutions including modifying Major to use a cross-attention algorithm to attend from one set of 2D data to the other two sets of 2D data.
Claim 14 is rejected under 35 U.S.C. 103 as being unpatentable over Major (B. Major et al., “Vehicle Detection With Automotive Radar Using Deep Learning on Range-Azimuth-Doppler Tensors,” 2019) and Lin (T. -Y. Lin et al., “Feature Pyramid Networks for Object Detection,” 2017), as applied to Claim 1 above, and further in view of Eschbaumer (US 2022/0342039).
Regarding Claim 14, Major teaches: wherein obtaining the 3D data comprises obtaining range data, antenna data, and doppler data, the computer-implemented method further comprising: processing at least one of a Fourier transformation algorithm or a dense layer algorithm ([Section 2]: “three-dimensional Fast Fourier Transformation”); …
Major does not explicitly teach – but Eschbaumer teaches: … processing an absolute number algorithm (Eschbaumer [0073]: “It should be noted that since an FFT output consists in general of complex values, a peak selection in an FFT output (such as the aggregate R/D map) may be understood as a selection based on absolute values (i.e., complex magnitudes of the complex outputs)”). It would have been obvious to one of ordinary skill in the art to modify Major and use an absolute number algorithm, as taught by Eschbaumer. Using an absolute number algorithm when processing FFT data is well-known in the art and is beneficial for reducing the complexity of the FFT data so that the data can be more easily used and understood.
Response to Arguments
Applicant’s arguments, see page 9, filed 07 July 2025, with respect to Specification Objections, Claim Objections, and Claim Rejection under 35 U.S.C. 112(b) have been fully considered and are persuasive. The objections and rejections have been overcome.
Applicant’s arguments, see pages 9-11, with respect to Claim Rejections under 35 U.S.C. 102 have been considered but are moot because the arguments do not apply to the specific combination of references being used in the current rejection. However, for clarity, specific arguments are addressed.
Applicant appears to argue that Major does not teach processing a position encoding algorithm on the second and third sets of 2D data. Examiner respectfully disagrees and asserts that Major teaches processing a position encoding algorithm on the first set of 2D data in Section 4.1. In section 4.1.2, the same processing is applied to all three sets of 2D data.
Conclusion
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 NOAH Y. ZHU whose telephone number is (571)270-0170. The examiner can normally be reached Monday-Friday, 8AM-4PM.
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, William J. Kelleher can be reached on (571) 272-7753. 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.
/NOAH YI MIN ZHU/Examiner, Art Unit 3648
/William Kelleher/Supervisory Patent Examiner, Art Unit 3648
Prosecution Insights