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 .
Information Disclosure Statement
The information disclosure statements (IDS) submitted on 06/26/2025 and 03/18/2024 are in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statements are being considered by the examiner.
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.
Claims 1-6, 8-13 and 15-20 are rejected under 35 U.S.C. 103 as being unpatentable over Fahim Mohammed et al (US 20220020156 A1) in view of Scott Cohen et al (US 20210027471 A1).
Regarding claim 1, Mohammed et al discloses a system (¶ [14]) comprising:
a processor (¶ [15]); and
a computer-readable medium storing instructions that are operative upon execution by the processor (¶ [15]) to:
receive a first image including an object (¶ [22]);
receive a first segmentation mask corresponding to an object of the first image (¶ [22]), wherein the first segmentation mask had been generated using [a first processing] (¶ [22]);
receive a second segmentation mask corresponding to the object of the first image (¶ [22]), wherein the second segmentation mask had been generated using a [second processing] different than the first [processing] (¶ [22]);
assign, by a quality predictor, a first quality score to the first segmentation mask (¶ [22]) without using ground truth for the first image (¶ [37]);
assign, by the quality predictor, a second quality score to the second segmentation mask (¶ [22]) without using ground truth for the first image (¶ [37]);
based on at least the first quality score exceeding the second quality score, select the first segmentation mask for an image processing task and not select the second segmentation mask for the image processing task (¶ [46]); and
perform the image processing task using the first segmentation mask (¶ [46]).
Mohammed et al fails to explicitly disclose generating the first and second segmentation masks using first and second object detectors.
Cohen et al, in the same field of endeavor of providing an object selection system that accurately detects and selects objects in a digital image (Abstract), teaches generating the first and second segmentation masks using first and second object detectors (¶ [50-51] and ¶ [296]).
It would have been obvious to one of ordinary skill in the art before the invention was effectively filed for the system as disclosed by Mohammed et al comprising a processor executing instructions to receive a first segmentation mask corresponding to an object of the first image, wherein the first segmentation mask had been generated using a first processing; receive a second segmentation mask corresponding to the object of the first image, wherein the second segmentation mask had been generated using a second processing different than the first processing to utilize the teachings of Cohen et al which teaches generating the first and second segmentation masks using first and second object detectors to improve detection and selection of objects based on object attributes.
Regarding claim 2, Mohammed et al discloses the system of claim 1 (see rejection of claim 1), wherein selecting the first segmentation mask for the image processing task comprises determining that the first quality score meets a quality threshold and wherein not selecting the second segmentation mask for the image processing task comprises determining that the second quality score does not meet the quality threshold (¶ [46]).
Regarding claim 3, Mohammed et al discloses the system of claim 1 (see rejection of claim 1), wherein assigning the first quality score and the second quality score is performed contemporaneously with performing the image processing task (¶ [18] assignment of the quality score in real-time).
Regarding claim 4, Mohammed et al discloses the system of claim 1 (see rejection of claim 1), wherein the instructions are further operative to:
generate the first segmentation mask using the first object detector (see rejection of claim 1); and generate the second segmentation mask using the second object detector (see rejection of claim 1).
Regarding claim 5, Mohammed discloses the system of claim 4 (see rejection of claim 4), wherein generating the first segmentation mask and the second segmentation mask is performed contemporaneously with assigning the first quality score and the second quality score (¶ [18]).
Regarding claim 6, Mohammed et al discloses the system of claim 1 (see rejection of claim 1), wherein the instructions are further operative to:
receive a plurality of training images, a plurality of segmentation masks corresponding to the plurality of training images, and a plurality of quality scores associated with each segmentation mask and training image (¶ [84]); and
using the plurality of training images, the plurality of segmentation masks, and the plurality of quality scores, train the quality predictor to assign quality scores to segmentation masks based on an input image, without needing ground truth for the input image (¶ [89]).
Regarding claim 8, Mohammed et al discloses a computer-implemented method (¶ [17]) comprising:
receiving a first image including an object (see rejection of claim 1);
receiving a first segmentation mask corresponding to an object of the first image, wherein the first segmentation mask had been generated using a first object detector (see rejection of claim 1);
receiving a second segmentation mask corresponding to the object of the first image, wherein the second segmentation mask had been generated using a second object detector different than the first object detector (see rejection of claim 1);
assigning, by a quality predictor, a first quality score to the first segmentation mask without using ground truth for the first image (see rejection of claim 1);
assigning, by the quality predictor, a second quality score to the second segmentation mask without using ground truth for the first image (see rejection of claim 1);
based on at least the first quality score exceeding the second quality score, selecting the first segmentation mask for an image processing task and not selecting the second segmentation mask for the image processing task (see rejection of claim 1); and
performing the image processing task using the first segmentation mask (see rejection of claim 1).
Regarding claim 9, Mohammed et al discloses the computer-implemented method of claim 8 (see rejection of claim 8), wherein selecting the first segmentation mask for the image processing task comprises determining that the first quality score meets a quality threshold and wherein not selecting the second segmentation mask for the image processing task comprises determining that the second quality score does not meet the quality threshold (see rejection of claim 2).
Regarding claim 10, Mohammed et al discloses the computer-implemented method of claim 8 (see rejection of claim 8), wherein assigning the first quality score and the second quality score is performed contemporaneously with performing the image processing task (see rejection of claim 3).
Regarding claim 11, Mohammed et al discloses the computer-implemented method of claim 8 (see rejection of claim 8), further comprising:
generating the first segmentation mask using the first object detector 8 (see rejection of claim 4); and
generating the second segmentation mask using the second object detector 8 (see rejection of claim 4).
Regarding claim 12, Mohammed et al discloses the computer-implemented method of claim 11 (see rejection of claim 1), wherein generating the first segmentation mask and the second segmentation mask is performed contemporaneously with assigning the first quality score and the second quality score (see rejection of claim 5).
Regarding claim 13, Mohammed et al discloses the computer-implemented method of claim 8 (see rejection of claim 8), further comprising:
receiving a plurality of training images, a plurality of segmentation masks corresponding to the plurality of training images, and a plurality of quality scores associated with each segmentation mask and training image (see rejection of claim 6); and
using the plurality of training images, the plurality of segmentation masks, and the plurality of quality scores, training the quality predictor to assign quality scores to segmentation masks based on an input image, without needing ground truth for the input image (see rejection of claim 6).
Regarding claim 15, Mohammed et al discloses a computer storage device having computer-executable instructions stored thereon, which, on execution by a computer, cause the computer to perform operations (see rejection of claim 1) comprising:
receiving a plurality of images and a plurality of segmentation masks each corresponding to an image of the plurality of images, wherein each segmentation mask of each plurality of segmentation masks had been generated using a different object detector or different object detector setting options (see rejection of claim 1);
assigning, by a quality predictor, a quality score to each segmentation mask without using ground truth (see rejection of claim 1);
determining a set of highest quality scores per each image of the plurality of segmentation masks, less than all of the quality scores per each image (see rejection of claim 1); and
performing an image processing task using segmentation masks having an assigned quality score within the set of highest quality scores (see rejection of claim 1).
Regarding claim 16, Mohammed et al discloses the computer storage device of claim 15 (see rejection of claim 15), wherein the set of highest quality scores comprises a plurality of quality scores (¶ [23]).
Regarding claim 17, Mohammed et al discloses the computer storage device of claim 15 (see rejection of claim 15), wherein the set of highest quality scores is a single quality score (¶ [46]).
Regarding claim 18, Mohammed et al discloses the computer storage device of claim 15 (see rejection of claim 15), wherein assigning the quality scores is performed contemporaneously with performing the image processing task (see rejection of claim 3).
Regarding claim 19, Mohammed et al discloses the computer storage device of claim 15 (see rejection of claim 15), wherein the operations further comprise:
generating the plurality of segmentation masks, wherein generating the plurality of segmentation masks is performed contemporaneously with assigning the quality scores (see rejection of claim 5).
Regarding claim 20, Mohammed et al discloses the computer storage device of claim 15 (see rejection of claim 15), wherein the operations further comprise:
receiving a plurality of training images, a plurality of segmentation masks corresponding to the plurality of training images, and a plurality of quality scores associated with each segmentation mask and training image (see rejection of claim 6); and
using the plurality of training images, the plurality of segmentation masks, and the plurality of quality scores, training the quality predictor to assign quality scores to segmentation masks based on an input image, without needing ground truth for the input image (see rejection of claim 6).
Claims 7 and 14 and rejected under 35 U.S.C. 103 as being unpatentable over Mohammed et al in view of Cohen et al as applied to claim 1 above, and further in view of Jong Yae Lee et al (US 20230206456 A1).
Regarding claim 7, Mohammed et al discloses the system of claim 1 (see rejection of claim 1).
Mohammed et al fails to explicitly disclose wherein the quality predictor comprises a multi-stage vision transformer model with a query head.
Lee et al, in the same field of endeavor of object segmentation from an input digital image (Abstract), teaches the quality predictor comprises a multi-stage vision transformer model with a query head (¶ [78]).
It would have been obvious to one of ordinary skill in the art before the invention was effectively filed for the system as disclosed by Mohammed et al comprising a processor executing instructions to receive a first segmentation mask corresponding to an object of the first image, wherein the first segmentation mask had been generated using a first processing; receive a second segmentation mask corresponding to the object of the first image, wherein the second segmentation mask had been generated using a second processing different than the first processing to utilize the teachings of Lee et al which teaches the quality predictor comprises a multi-stage vision transformer model with a query to improve the detection of boundaries of objects to enhance accuracy in object segmentation.
Regarding claim 14, Mohammed et al discloses the computer-implemented method of claim 8 (see rejection of claim 8), wherein the quality predictor comprises a multi-stage vision transformer model with a query head (see rejection of claim 7).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JAMARES Q WASHINGTON whose telephone number is (571) 270-1585. The examiner can normally be reached Mon-Fri 8:30am-4:30pm.
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, Akwasi M. Sarpong can be reached at (571) 270-3438. 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.
/JAMARES Q WASHINGTON/Primary Examiner, Art Unit 2681
December 23, 2025