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 .
Claim Interpretation
Claim 14 is considered as independent claims because it simply refers to other claim as a matter of short-hand drafting technique.
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.
Claim 14 is rejected under 35 U.S.C. 101 because the claimed invention is directed to non-statutory subject matter. The claim(s) does/do not fall within at least one of the four categories of patent eligible subject matter because claim 14 is directed to a “computer-readable storage medium". However, according to paragraph [0127] of the specification, the broadest reasonable interpretation of the "computer-readable storage medium" covers a transitory propagating signal which is non-statutory subject matter. See In re Nuijten, 500 F.3d 1346, 84 USPQ2d 1495 (Fed. Cir. 2007).
The examiner suggests amending the claim(s) to recite a “non-transitory computer-readable medium” storing a computer program or equivalent. Any amendment to the claims should be commensurate with its corresponding disclosure.
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.
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.
Claim(s) 1, 10, 12-15, 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Tian et al. [“Efficient and Hybrid Decoder for Local Map Construction in Bird'-Eye-View”], hereinafter Tian, in view of Yu et al. [“ScalableMap: Scalable Map Learning for Online Long-Range Vectorized HD Map Construction”], hereinafter Yu.
As to claim 1, Tian discloses a method with map construction, comprising:
extracting a bird's-eye view (BEV) feature map based on input data (Section III.A., page 8380, e.g., “First, three stacked convolutional blocks further aggregate the original BEV features output by the encoder, thereby enhancing the local association in the BEV coordinate system. The serial feature extraction process reduces the resolution of feature maps to 1/2, 1/4, and 1/8 of the original input. Second, we upsample and fuse low-resolution feature representations at different semantic levels, which aims to model the context information in BEV. This design is inspired by the feature pyramid
network (FPN) [40], but has a simpler structure. Considering the computation and inference efficiency of the framework, only one upsampling unit is adopted. After that, the aggregated BEV features are fed into task-specific decoding modules, namely segmentation head (Seg head) and query-based
lane decoder (QLD)”):
determining map information through a hybrid decoder based on the BEV feature map and a hybrid query (Figs. 2, 3, page 8380, e.g., Efficient and Hybrid Decoder); and
constructing a map corresponding to the input data based on the map information (Section III.A, B, pages 8379, 8380, e.g., “we disassemble the local map construction process into two parts, i.e., encoder and decoder, so that a modular configuration design paradigm is adopted in the specific implementation; and ”Following the setting of HDMapNet, H and W are set to 200 and 400”);
wherein the map comprises a plurality of map elements (page 8380, e.g., “The channel
dimension is 4, corresponding to three kinds of lanes (divider, pedestrian crossing, boundary) and the background category”);
wherein the map information comprises coordinate information and class information of the plurality of map elements (pages 8379, 8380, e.g., “The function of BEV encoder is to project and aggregate the outputs of different camera coordinate systems into a common BEV coordinate system”, and “The channel dimension is 4, corresponding to three kinds of lanes (divider, pedestrian crossing, boundary) and the background category”);
wherein each of the plurality of map elements comprises an area formed by a plurality of coordinate points in the map (pages 8379, 8380, e.g., “The function of BEV encoder is to project and aggregate the outputs of different camera coordinate systems into a common BEV coordinate system”, and “The channel dimension is 4, corresponding to three kinds of lanes (divider, pedestrian crossing, boundary) and the background category”);
wherein the hybrid query comprises a plurality of hybrid features (pages 8380, 8381, “Figure 3 (b) shows the structure of query-based decoder, whose working pattern consists of two steps. First, defining the initial query embeddings (IQ), which are used to store the statistics of the latent lane instances. The first Transformer decoder updates IQ based on the BEV features of each timestamp. This preliminary decoding can be viewed as a refinement of the initial query embeddings. After that, the output queries are highly correlated with dynamic inputs, e.g., images from surrounding cameras. The second Transformer decoder queries the aggregated BEV features with the positive embeddings (PQ) that are assigned with ground truth lanes (GTs), and outputs the corresponding instance level perception results”),
wherein each of the plurality of hybrid features comprises a point feature and an element feature corresponding to a map element (pages 8380, 8381, e.g., “The channel dimension is 4, corresponding to three kinds of lanes (divider, pedestrian crossing, boundary) and the background category”, and “Figure 3 (b) shows the structure of query-based decoder, whose working pattern consists of two steps. First, defining the initial query embeddings (IQ), which are used to store the statistics of the latent lane instances. The first Transformer decoder updates IQ based on the BEV features of each timestamp. This preliminary decoding can be viewed as a refinement of the initial query embeddings. After that, the output queries are highly correlated with dynamic inputs, e.g., images from surrounding cameras. The second Transformer decoder queries the aggregated BEV features with the positive embeddings (PQ) that are assigned with ground truth lanes (GTs), and outputs the corresponding instance level perception results”);
wherein the point feature represents information associated with each coordinate point of the map element (pages 8380, 8381, e.g., “The function of BEV encoder is to project and aggregate the outputs of different camera coordinate systems into a common BEV coordinate system”, and “The channel dimension is 4, corresponding to three kinds of lanes (divider, pedestrian crossing, boundary) and the background category”, “the class label in one-hot form, and K = H × W is the number of pixel features in the BEV coordinate system”, “Figure 3 (b) shows the structure of query-based decoder, whose working pattern consists of two steps. First, defining the initial query embeddings (IQ), which are used to store the statistics of the latent lane instances. The first Transformer decoder updates IQ based on the BEV features of each timestamp. This preliminary decoding can be viewed as a refinement of the initial query embeddings. After that, the output queries are highly correlated with dynamic inputs, e.g., images from surrounding cameras. The second Transformer decoder queries the aggregated BEV features with the positive embeddings (PQ) that are assigned with ground truth lanes (GTs), and outputs the corresponding instance level perception results”), and
wherein the element feature represents information associated with the map element (pages 8380, 8381, e.g., “The function of BEV encoder is to project and aggregate the outputs of different camera coordinate systems into a common BEV coordinate system”, and “The channel dimension is 4, corresponding to three kinds of lanes (divider, pedestrian crossing, boundary) and the background category”, “the class label in one-hot form, and K = H × W is the number of pixel features in the BEV coordinate system”, “Figure 3 (b) shows the structure of query-based decoder, whose working pattern consists of two steps. First, defining the initial query embeddings (IQ), which are used to store the statistics of the latent lane instances. The first Transformer decoder updates IQ based on the BEV features of each timestamp. This preliminary decoding can be viewed as a refinement of the initial query embeddings. After that, the output queries are highly correlated with dynamic inputs, e.g., images from surrounding cameras. The second Transformer decoder queries the aggregated BEV features with the positive embeddings (PQ) that are assigned with ground truth lanes (GTs), and outputs the corresponding instance level perception results”).
Tian does not explicitly describe the local map construction as a high-definition (HD) map construction.
Yu teaches constructing a high-definition (HD) map (Abstract, “We propose a novel end-to-end pipeline for online long-range vectorized high-definition (HD) map construction using on-board camera sensors”).
It would have been obvious to one of ordinary skill in the art to incorporate Yu’s teachings into Tian since doing so would merely combine prior art elements according to known methods to yield predictable results, and improve the performance of map construction.
As to claim 10, the combination of Tian and Yu discloses the method of claim 1, wherein a loss function used by the hybrid decoder during a training process comprises a point-element consistency loss, wherein the point-element consistency loss is used to represent a level of risk of inconsistency between a point query and an element query of the updated hybrid query (Tian, page 8380, 8381, Fig. 2, e.g., Seg. Loss, Query Loss, “The final output and training labels in the BEV coordinate system are used to compute segmentation loss and detection loss, supervising the update of model parameters”, “The segmentation loss is calculated using the ground truth and classification map activated by softmax function”, “Each IQ will calculate the classification loss of C categories”).
As to claim 12, the combination of Tian and Yu discloses the method of claim 10, wherein the loss function used by the hybrid decoder during the training process further comprises at least one of a semantic segmentation loss, a classification loss, a point regression loss, a point orientation loss, or a mask loss (Tian, page 8380, 8381, Fig. 2, e.g., Seg. Loss, Query Loss, “The final output and training labels in the BEV coordinate system are used to compute segmentation loss and detection loss, supervising the update of model parameters”, “The segmentation loss is calculated using the ground truth and classification map activated by softmax function”, “Each IQ will calculate the classification loss of C categories”).
As to claim 15, the combination of Tian and Yu discloses the method of claim 1, further comprising: using at least one sensor to collect sensor data as the input data (Tian, Section III.A., page 8380, e.g., “The raw input of our method can be surrounding images and/or point clouds captured by on-board sensors”).
As to claims 13, 14, 20, these claims recite features similar to those discussed above. Therefore, they are rejected for reasons similar to those discussed above.
Allowable Subject Matter
Claims 2-9, 11, 16-19 are objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
The following is a statement of reasons for the indication of allowable subject matter: The prior art discloses the claim limitations discussed above, but fails to disclose the combined features required by each of dependent claims 2, 11, 16.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Yang et al. disclose techniques for improving the performance of an autonomous vehicle (AV) by automatically annotating objects surrounding the AV.
Wang et al. relate to generating a synthetic high-definition (HD) map from a standard-definition (SD) map, and simulating changes in the synthetic HD map.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to PHUOC TRAN whose telephone number is (571)272-7399. The examiner can normally be reached 9am-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, Vu Le can be reached at 571-272-7332. 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.
/PHUOC TRAN/Primary Examiner, Art Unit 2668
Prosecution Insights