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 .
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 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.
Claims 1-20 are rejected under 35 U.S.C. 103 as being unpatentable over D11.
With regard to claim 1, D1 teach computer-implemented method of performing post processing on data generated by processing an image in accordance with a single-shot detector (SSD) neural network, the data comprising information identifying a plurality of bounding boxes in the image and a confidence score for a class for each bounding box of the plurality of bounding boxes (see ¶ 1: single shot detection; see figs. 1, 3: bounding boxes and corresponding confidence scores); for each bounding box of the plurality of bounding boxes: (a) determining if the confidence score for the bounding box meets a confidence score threshold (see fig. 3: step 370 compares the confidence score to a threshold; see also ¶ 119);
see ¶¶ 115-117: outputting a sorted list of 100 or 200 most confident detection boxes).
D1 teach obtaining detection box data including confidence scores and outputting a list of 100 or 200 most confident detection boxes, but fails to explicitly teach the sequential steps generating and updating a list of 100 or 200 most confident detection boxes. However, the steps described in the claim, i.e., checking if less than a maximum number (in this case 100 or 200) of box entries have been stored, if less than a maximum number are stored adding a new bounding box entry, and if a maximum have been stored, then replacing the least confident box with the new box being processed. However, Examiner takes Official Notice to the fact that these steps for generating and updating a fixed length list are extremely well known in the art before the effective filing date and would have been particularly obvious for one skilled in the art to incorporate known teachings into the configuration of D1 yielding predictable and enhanced results. The motivation for dynamically updating a list of 100 or 200 most confident boxes would have been to enhance the efficiency of processing detection bounding boxes.
With regard to claim 2, D1 fails to explicitly teach in response to determining that the confidence score for the bounding box is greater than the lowest confidence score of the bounding box entries, (i) identifying a new lowest confidence score in the bounding box entries after the bounding box entry with the lowest confidence score has been removed and the new bounding box entry has been added and (ii) storing the new lowest confidence score for use in performing (d) for a subsequent bounding box of the plurality of bounding boxes. D1 teaches sorting the list with a first entry being the most confident score and the last entry with the least confidence score. However, Examiner takes Official Notice to the fact that these steps are extremely well known in the art before the effective filing date, i.e., it is well known in the art to update a sorted list by identifying the entry with the lowest score to be replaced with a new entry satisfying a threshold score and which is greater than the lowest score entry. It would have been obvious for one skilled to incorporate known teachings into the configuration of D1 yielding predictable and enhanced updating of a sorted list.
With regard to claim 3, D1 teach wherein the bounding box entries are stored in an array with a fixed number of elements (see ¶¶ 115-117: list or array of the 100 or 200 most confident detection boxes, implicit that the list is a fixed array).
With regard to claim 4, D1 teach wherein the bounding box entries are stored in an ordered data structure in which a position of the bounding box entries in the ordered data structure is based on the confidence scores of the bounding box entries (see ¶¶ 115-117: sorting the list based on confidence score).
With regard to claim 5, D1 teach wherein removing the bounding box entry with the lowest confidence score and adding a new bounding box entry for the bounding box comprises replacing the bounding box entry with the lowest confidence score in memory with the new bounding box entry for the bounding box (see ¶¶ 115-117: maintaining a list of 100 or 200 most confident detection boxes implicitly means that the lowest confidence entry with a new entry).
With regard to claim 6, D1 fails to explicitly teach wherein the bounding box entries are stored in a binary search tree structure. However, Examiner takes Official Notice to the fact that binary search tree structures are extremely well known in the art before the effective filing date and one skilled in the art would have been motivated to incorporate known teachings into the configuration of D1 yielding predictable and enhanced results including reducing the time required to search, insert or delete elements.
With regard to claim 7, D1 fails to explicitly teach further comprising storing, for each bounding box entry, a node that comprises a pointer to a left child node, a pointer to a right child node and information identifying the associated bounding box entry. However, Examiner takes Official Notice to the fact that binary search tree structures comprising left and right nodes are extremely well known in the art before the effective filing date and one skilled in the art would have been motivated to incorporate known teachings into the configuration of D1 yielding predictable and enhanced results including reducing the time required to search, insert or delete elements.
With regard to claim 8, D1 teach wherein, when the number of bounding box entries is less than the maximum number of bounding box entries, the bounding box entries are stored in an unsorted structure (see ¶¶ 115-117: list maybe left unsorted). D1 fails to explicitly teach when the number of bounding box entries is equal to the maximum number, the bounding box entries are stored in a min heap structure. However, Examiner takes Official Notice to the fact that min heap structures are extremely well known in the art before the effective filing date and one skilled in the art would have been motivated to incorporate known teachings into the configuration of D1 yielding predictable and enhanced results. The motivation would have been for fast access to the absolute smallest element, such as a bounding box with the lowest confidence score, for example.
With regard to claim 9, D1 fails to explicitly teach wherein the bounding box entries are stored in a min heap structure. However, Examiner takes Official Notice to the fact that min heap structures are extremely well known in the art before the effective filing date and one skilled in the art would have been motivated to incorporate known teachings into the configuration of D1 yielding predictable and enhanced results. The motivation would have been for fast access to the absolute smallest element, such as a bounding box with the lowest confidence score, for example.
With regard to claim 10, D1 fails to explicitly teach wherein the bounding box entries are stored in a singly linked list structure, further comprising storing, for each bounding box entry, a node that comprises a pointer to a node associated with a bounding box entry with a next highest confidence score or a next lowest confidence score, and information identifying the bounding box entry associated with the node. However, Examiner takes Official Notice to the fact that singly linked structures are extremely well known in the art before the effective filing date and one skilled in the art would have been motivated to incorporate known teachings into the configuration of D1 yielding predictable and enhanced results. The motivation would have been to store entries in an ascending or descending order where each element (node) contains a data value and a pointer to the next node.
With regard to claim 11, D1 teach wherein each bounding box is associated with a unique bounding box identifier and the information in a bounding box entry that identifies a bounding box of the plurality of bounding boxes comprises the unique bounding box identifier for that bounding box (see ¶ 100: sort index to sort the detection boxes, each entry in the sort index identifies a different detection box, e.g. using a value or pointer, where the value or pointer is read as an identifier).
With regard to claim 12, D1 teach wherein the information in a bounding box entry that identifies a bounding box comprises parameters of that bounding box (see ¶ 52: co-ordinates of a bounding box).
With regard to claim 13, D1 teach wherein the information in a bounding box entry that identifies a bounding box comprises co-ordinates of that bounding box (see ¶ 52: co-ordinates of a bounding box).
With regard to claim 14, D1 fails to explicitly teach wherein the method further comprises, prior to adding a new bounding box entry for a bounding box, determining parameters of the bounding box from parameters of an anchor box and a set of one or more deltas. However, Examiner takes Official Notice to the fact that anchor boxes are extremely well known in the art before the effective filing date and one skilled in the art would have been motivated to incorporate known teachings into the configuration of D1 yielding predictable and enhanced results. The motivation for using anchor boxes is that they help computer vision model locate and classify objects by using them as starting points.
With regard to claim 15, D1 teach herein the parameters of the bounding box comprise x and y co-ordinates see ¶ 52: co-ordinates indicating the relative location of an upper left vertex of the detection box in the image and dimension information indicating the height and width of the detection box). D1 fails to explicitly teach using coordinates of a center of the bounding box, however, it would have been obvious for one skilled in the art to uniquely identify the position of the bounding box by using coordinates of the middle or center of the bounding box instead of the vertices, for example.
With regard to claim 16, D1 teach wherein, for each bounding box entry, the information identifying a bounding box and the confidence score are stored in memory such that the information identifying the bounding box is separately accessible from the confidence score (see ¶¶ 50-52: coordinates of the bounding box and confidence value). It would have been obvious for one skilled in the art to associate the information of the bounding box such as coordinates and the confidence score and store them in separate locations in the memory for efficiency.
With regard to claim 17, D1 teach further comprising performing non-maximum suppression on the sorted list of bounding box entries (see abstract, ¶¶ 97, 181: non-maximum suppression on the plurality of detection boxes).
With regard to claim 18, D1 teach wherein the one or more processors comprises a micro-controller, wherein the micro-controller is embedded in a neural network accelerator, wherein the neural network accelerator comprises one or more hardware accelerators for performing neural network operations and the method further comprises processing the image in accordance with the SSD neural network using the one or more hardware accelerators of the NNA to generate the data (see ¶¶ 28, 252, 255-256: neural network accelerator, hardware accelerators etc.).
With regard to claims 19-20, see discussion of claim 1.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to AVINASH YENTRAPATI whose telephone number is (571)270-7982. The examiner can normally be reached on 8AM-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, Sumati Lefkowitz can be reached on (571) 272-3638. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/AVINASH YENTRAPATI/Primary Examiner, Art Unit 2672
1 US Publication No. 2021/0350161.
Prosecution Insights