Prosecution Insights
Last updated: November 04, 2026
Application No. 17/948,124

MULTI-SENSOR OCCLUSION-AWARE TRACKING OF OBJECTS IN TRAFFIC MONITORING SYSTEMS AND METHODS

Non-Final OA §103
Filed
Sep 19, 2022
Priority
Mar 25, 2020 — provisional 62/994,709 +1 more
Examiner
KOZIOL, STEPHEN R
Art Unit
2665
Tech Center
2600 — Communications
Assignee
Flir Systems Trading Belgium Bvba
OA Round
3 (Non-Final)
78%
Grant Probability
Favorable
3-4
OA Rounds
0m
Est. Remaining
99%
With Interview

Examiner Intelligence

Grants 78% — above average
78%
Career Allowance Rate
309 granted / 398 resolved
+15.6% vs TC avg
Strong +25% interview lift
Without
With
+25.1%
Interview Lift
resolved cases with interview
Typical timeline
3y 9m
Avg Prosecution
6 currently pending
Career history
409
Total Applications
across all art units

Statute-Specific Performance

§101
23.2%
-16.8% vs TC avg
§103
39.7%
-0.3% vs TC avg
§102
17.3%
-22.7% vs TC avg
§112
17.3%
-22.7% vs TC avg
Black line = Tech Center average estimate • Based on career data from 398 resolved cases

Office Action

§103
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 statement(s) (IDS) submitted on 02/28/2023, 08/22/2024, and 08/04/2025 was/were filed. The submission(s) is/are in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement(s) is/are being considered by the examiner. Response to Arguments The restriction of claims 1-6 are withdrawn in light of the amendments. All drawing objections, claim interpretation under 35 U.S.C. § 112(f), and claim rejections under 35 U.S.C. § 101 and 35 U.S.C. § 112(b) are withdrawn. Applicant’s arguments with respect to claim(s) 7-20 rejected under 35 U.S.C. §§ 102, 103 have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument. Claim Objections Claim 9 objected to because of the following informalities: The claim recites “The system of claim 7, wherein wherein”. Appropriate correction is required. Claim Interpretation Note that according to the Federal Circuit’s 2004 Superguide v. DirecTV decision, “at least one of … and …” requires at least one instance of each and every item listed. Claims 8 and 16 recite “at least two of a visual image sensor, a thermal image sensor and a radar sensor”. If Applicant intends for an interpretation of only one of these items being required for claim interpretation, Applicant can amend the claim language to, instead recite, “at least one of … or …”. In SuperGuide, the Federal Circuit held that the plain meaning of “at least one of A, B, and C” means: at least one of A, at least one of B, and at least one of C. The Court held that if the applicant intended “at least one of A, B, and C” to mean A, B, or C, they should have used “OR.” The specification shows that the meaning of “and” is meant to connotate a disjunctive “or” and will be treated as a disjunctive “or” for purposes of the BRI of the claim in light of the specification (e.g., “at least one of x, or at least one of y, or at least one of z”). Claim Rejections - 35 USC § 103 Claims 1, 7-12, 14-18, 20, 21, and 24 are rejected under 35 U.S.C. 103 as being unpatentable over Liu et al. (US 20190258876 A1), hereinafter referred to as Liu, in view of Silva et al. (US 20200225669 A1), hereinafter referred to as Silva. Regarding claim 7, Liu teaches a system comprising: a plurality of sensors configured to capture data associated with a traffic location (Liu, ¶ 2, “An autonomous or semi-autonomous vehicle includes various sensors that provide multiple forms of data regarding the objects in the environmental surroundings of the vehicle. One set of data can be used to provide a three-dimensional point cloud of the environment. Another set of data can be used to perform semantic segmentation of the environment from a two-dimensional image, allowing the vehicle to identify objects in the environment.”; teaches various sensors that capture data (images) associated with surroundings of the vehicle, which would be traffic locations) ; a computer system (Liu, ¶ 29, “The controller 34 includes at least one processor 44 and a computer readable storage device or media 46. The processor 44 can be any custom made or commercially available processor, a central processing unit (CPU), a graphics processing unit (GPU), an auxiliary processor among several processors associated with the controller 34, a semiconductor based microprocessor (in the form of a microchip or chip set), a macroprocessor, any combination thereof, or generally any device for executing instructions.”; teaches a combination of devices for executing instructions and computations) configured to: detect one or more objects in the captured data (Liu, ¶ 34, “A first branch (including boxes 202, 204, 206 and 208) includes methods for detecting and classifying objects in a two-dimensional image obtained from a digital camera of the vehicle.”; teaches detecting objects from data (images) captured by a camera); determine an object location within the captured data (Liu, ¶ 35, “Using the features determined at box 204, object detection can be performed at box 206 by determining bounding boxes for objects in the image. The features determined at box 204 can also be used to perform semantic segmentation of the digital image at box 208 in order to classify objects in the 2-D digital image.”; teaches locating an object within a captured image); transform each object location to world coordinates associated with one of the plurality of sensors (Liu, ¶ 34, “A second branch (including boxes 210, 212, 214, and 216) includes methods for obtaining a three-dimensional point cloud of the surrounding environment. The coordinates of points in the point cloud is determined by the locations of the objects or targets within the environment for the second branch. […] In the second branch, in box 210 a 3D point cloud is received from various sensors that can include a digital camera, a pair of digital cameras arranged for stereoscopic imaging, Lidar, radar, etc. The sensor can further provide video data or temporally separated images.”; teaches determining coordinates of points based on the location of objects from sensors); and track each object location using the world coordinates (Liu, ¶ 44, “Returning to the first module 502, the first module compares points with a previous frame 506 to points within a current frame 508 in order to determine common movement of points or point clusters. Such commonly moving points can be used to determine a 3-D template of an object within the 3D point cloud.”; teaches tracking object movement using the three-dimensional point cloud points); Liu fails to teach wherein the tracking each object location using the world coordinates comprises predicting occlusions; wherein the predicting occlusions comprises, for at least a first object and a second object which are two of the one or more detected objects: estimating a height of at least one of the first and second objects; and predicting an occlusion of the second object by the first object based on the estimated height and on distances of the first and second objects from at least one of the sensors. Silva teaches wherein the tracking each object location using the world coordinates comprises predicting occlusions (Silva, ¶ 13, “This disclosure is directed to evaluating trajectories based on predicted occluded regions associated with the trajectories, where the predicted occluded regions are determined with respect to predicted locations of objects in an environment. For example, sensors of a vehicle (such as an autonomous vehicle) can capture sensor data of an environment, which may include objects separate from the vehicle, such as other vehicles, pedestrians, buildings, vegetation, and so forth. Sensor coverage of the environment may be limited (occluded) based on obstructions in the environment. A planning system associated with the vehicle can generate trajectories for the vehicle to follow in the environment to navigate around such objects. An occlusion system associated with the vehicle can determine predicted occluded regions associated with objects based on the trajectories for the vehicle to follow in the environment and/or predicted trajectories associated with the detected objects.”; teaches determining predicted occluded regions based on trajectories of a vehicle); wherein the predicting occlusions comprises, for at least a first object and a second object which are two of the one or more detected objects (Silva, FIG. 5A: PNG media_image1.png 647 984 media_image1.png Greyscale ¶ 66-67, “FIG. 5A is an overview of an example 500 presented in FIGS. 5B and 5C. An environment 502 may include a vehicle 504, a first object 506, and a second object 508. In this example, the first object 506 may be a truck that has parked on a drivable surface 510 and has its hazard lights on, such as to make a delivery nearby (e.g., the object 506 may represent a double-parked vehicle). Additionally, in this example, the second object 508 may be a vehicle moving in a same direction as the vehicle 504. The drivable surface 510 may be a two-lane road, with traffic moving in opposite directions in the two lanes, and accordingly, the object 508 can be associated with a predicted trajectory 509 that navigates around the object 506. In some examples, a planning system of the vehicle 504 may be evaluating a first trajectory 512 and a second trajectory 514 with respect to predicted location(s) of the object 506 and the object 508 over time. As illustrated, the environment 502 may represent an initial time (e.g., T1), at which the vehicle 504 is approaching the stationary object 506, and the object 508 is passing the object 506 in the oncoming traffic lane of the drivable surface 510. ”; teaches detecting and tracking two objects over time): estimating a height of at least one of the first and second objects (Silva, ¶ 33, “Although generally depicted as being in an x-direction and a y-direction for clarity in the example 124, the predicted occluded region 126 may also represent occlusion information associated with a z-direction component, which may correspond to a height of the object 106, for instance. In another example, occlusion information may be associated with a z-direction component corresponding to terrain of the surrounding environment, such as hills or valleys, or height of objects that may be occluded but are relevant to trajectory determinations, such as stop lights or train crossing signals.”; teaches the predicted occluded region having information about the z-direction component or height of objects); and predicting an occlusion of the second object by the first object based on the estimated height and on distances of the first and second objects from at least one of the sensors (Silva, ¶ 33, “Although generally depicted as being in an x-direction and a y-direction for clarity in the example 124, the predicted occluded region 126 may also represent occlusion information associated with a z-direction component, which may correspond to a height of the object 106, for instance. In another example, occlusion information may be associated with a z-direction component corresponding to terrain of the surrounding environment, such as hills or valleys, or height of objects that may be occluded but are relevant to trajectory determinations, such as stop lights or train crossing signals.”; teaches the predicted occluded region having information about the x-, y-, and z-direction component of objects; FIG. 5B: PNG media_image2.png 987 607 media_image2.png Greyscale ¶ 69, “For instance, FIG. 5B is an example 524 of evaluating the first trajectory 512 of the vehicle 504 based on predicted occluded region(s) associated with object(s) over time, in accordance with embodiments of the disclosure. The environment 516 is reproduced for convenience, depicting the first trajectory 512 that the vehicle 504 may follow. A predicted environment 526 illustrates the vehicle 504, the object 506, and the object 508 at a future time (e.g., at T2A, a time after T1), where the vehicle 504 follows the first trajectory 512. Additionally, a predicted environment 528 illustrates the vehicle 504, the object 506, and the object 508 at a subsequent future time (e.g., at T3A, a time after T2A), where the vehicle 504 continues along the first trajectory 512.”; teaches predicting where the objects will be based on trajectory, which involves distance calculation). It would have been obvious to a person having ordinary skill in the art before the time of the effective filing date of the claimed invention of the instant application to implement the occlusion region prediction as taught by Silva using the hardware system and sensors of Liu. The suggestion/motivation for doing so would have been that “the safety of the vehicle can be improved by allowing the vehicle to adapt and react to its surroundings, rather than requiring the vehicle to follow set instructions, especially when a situation arises in which there is not a specific rule present. Further, controlling the vehicle based in part on predicted occluded regions can reduce processing resources, as the number of enumerated rules for every driving situation the vehicle may encounter would be virtually infinite. Accordingly, techniques for analyzing trajectories based on predicted occluded regions can be performed faster than conventional techniques, which may allow for a faster response or may allow a computing system to consider additional alternative trajectories, thereby improving safety outcomes, performance, and/or accuracy” (Silva, ¶ 21). Further, one skilled in the art could have combined the elements as described above by known methods with no change in their respective functions, and the combination would have yielded nothing more than predictable results. Therefore, it would have been obvious to combine Liu with Silva to obtain the invention as specified in claim 7. Regarding claim 8, Liu in view of Silva teaches the system of claim 7, wherein the plurality of sensors comprises at least two sensors from a list of: (a visual image sensor, a thermal image sensor and a radar sensor) (Liu, ¶ 28, “The sensor system 28 includes one or more sensing devices 40a-40n that sense observable conditions of the exterior environment and/or the interior environment of the autonomous vehicle 10. The sensing devices 40a-40n can include, but are not limited to, radars, lidars, global positioning systems, optical cameras, thermal cameras, ultrasonic sensors, and/or other sensors. The cameras can include two or more digital cameras spaced at a selected distance from each other, in which the two or more digital cameras are used to obtain stereoscopic images of the surrounding environment in order to obtain a three-dimensional image.”; teaches a visual image sensor (optical camera), thermal camera, and radar sensor). Regarding claim 9, Liu in view of Silva teaches the system of claim 7, wherein wherein the determining an object location within the captured data comprises an object localization process comprising a trained deep learning process (Liu, ¶ 35, “In the first branch, box 202 includes training sets images that are provided to train a convolutional neural network (CNN). Once trained, in box 204 the CNN can be used on two-dimensional images obtained at the vehicle to perform feature extraction from the 2-D images in order to identify objects in the 2-D images. Using the features determined at box 204, object detection can be performed at box 206 by determining bounding boxes for objects in the image.”; teaches a CNN that can identify objects in 2-D images and determine bounding boxes around them). Regarding claim 10, Liu in view of Silva teaches the system of claim 9, wherein the deep learning process is configured to receive the captured data from one of the sensors and determine a bounding box surrounding the detected object (Liu, ¶ 35, “In the first branch, box 202 includes training sets images that are provided to train a convolutional neural network (CNN). Once trained, in box 204 the CNN can be used on two-dimensional images obtained at the vehicle to perform feature extraction from the 2-D images in order to identify objects in the 2-D images. Using the features determined at box 204, object detection can be performed at box 206 by determining bounding boxes for objects in the image.”; teaches a CNN that receives images obtained from sensors and puts a bounding box around the object). Regarding claim 11, Liu in view of Silva teaches the system of claim 9, wherein the deep learning process is configured to receive the captured data from one of the sensors, detect an object in the captured data and output a classification of the detected object including a confidence factor (Liu, ¶ 38, “Semantic segmentation groups pixels of the image according to separate classes, such as roadway, vehicle, traffic light, etc. Semantic segmentation enables classification of the objects in the image and also associates the classification of the objects with relevant bounding boxes. In one embodiment, the convolutional neural network (CNN) predicts the object's bounding box 310 and its class probability 312 simultaneously.”; teaches the CNN detecting an object, a class of the object, and the class probability (confidence)). Regarding claim 12, Liu in view of Silva teaches the system of claim 7, wherein the computer system is further configured to perform a distance matching algorithm comprising synthesizing the detected objects detected in the data captured from the plurality of sensors (Liu, ¶ 36, “In box 212, the processor extracts various features from the selected point cloud. In box 214, the extracted features are compared to templates of objects in order to match the extracted feature to a selected template. For example, a selected set of points forming an octagonal shape can be determined from the 3D point cloud and the octagonal shape can be matched to a template for a STOP sign in order to identify the sign.”; teaches the processor that performs matching between points and templates to form and identify the object. Matching a shape would also match distances between the points.). Regarding claim 14, Liu in view of Silva teaches the system of claim 7, wherein the computer system is further configured to transform the tracked objects to three-dimensional objects in the world coordinates (Liu, ¶ 44, “Returning to the first module 502, the first module compares points with a previous frame 506 to points within a current frame 508 in order to determine common movement of points or point clusters. Such commonly moving points can be used to determine a 3-D template of an object within the 3D point cloud.” […] Once a cluster of points is identified, a representative coordinate 512 of the cluster can be provided to the second module 504. The second module 504 receives the representative coordinate 512 and provides a bounding box 514 that includes or is associated with the representative coordinate to the first module 504.; teaches transforming the 3D point cloud of the objects to a 3-D template of the object which are associated with the coordinates). Regarding claim 21, Liu in view of Silva teaches the system of claim 7, wherein the predicting an occlusion based on the estimated height and on distances of the first and second objects from at least one of the sensors comprises determining a potential occlusion area of the first object, the potential occlusion area being an area in which another object may be occluded by the first object (Silva, FIG. 5B: PNG media_image2.png 987 607 media_image2.png Greyscale ¶ 69, “For instance, FIG. 5B is an example 524 of evaluating the first trajectory 512 of the vehicle 504 based on predicted occluded region(s) associated with object(s) over time, in accordance with embodiments of the disclosure. The environment 516 is reproduced for convenience, depicting the first trajectory 512 that the vehicle 504 may follow. A predicted environment 526 illustrates the vehicle 504, the object 506, and the object 508 at a future time (e.g., at T2A, a time after T1), where the vehicle 504 follows the first trajectory 512. Additionally, a predicted environment 528 illustrates the vehicle 504, the object 506, and the object 508 at a subsequent future time (e.g., at T3A, a time after T2A), where the vehicle 504 continues along the first trajectory 512.”; teaches predicted occluded region 534 from object 506 and object 508 being occluded by it). Regarding claim 1, Liu in view of Silva teaches the system of claim 7, wherein the at least one of the sensors is an image sensor configured to capture a stream of images of a scene (Liu, ¶ 36, “In the second branch, in box 210 a 3D point cloud is received from various sensors that can include a digital camera, a pair of digital cameras arranged for stereoscopic imaging, Lidar, radar, etc. The sensor can further provide video data or temporally separated images.”; teaches sensors, such as a camera, that can capture video). Regarding claims 15, 16, 17, 18, 20, and 24 arguments made in rejecting claims 7, 8 with 12, 9 with 10, 11, 14, and 21 are analogous, respectively. Claims 22 and 25 are rejected under 35 U.S.C. 103 as being unpatentable over Liu in view of Silva as applied to claims 1, 7-12, 14-18, 20, 21, and 24 above, and further in view of Kobilarov (US11126180B1). Regarding claim 22, Liu in view of Silva teaches the system of claim 7. Liu in view of Silva fails to teach further comprising, if the occlusion occurs: tracking the second object, wherein the tracking the second object comprises predicting the second object location; and continuing to track the second object after the first object moves away not to occlude the second object. Kobilarov teaches further comprising, if the occlusion occurs: tracking the second object, wherein the tracking the second object comprises predicting the second object location (Kobilarov, FIG. 4: PNG media_image3.png 666 988 media_image3.png Greyscale Col. 11 lines 22-48, “In some examples, the attribute(s) 402 may include, but are not limited to, occluding object information (e.g., a bounding box, a classification, a location, a velocity, a volume, an indication that the object is a double-parked vehicle, lighting states (e.g., brake light(s), blinker light(s), hazard light(s), headlight(s), reverse light(s), etc.), a wheel orientation of the object, etc.), occlusion information (e.g., size, shape, location, etc.), vehicle distance to the occlusion (e.g., a distance from the vehicle 408 to a point associated with the occluded region 418), object distances(s) to the occlusion (e.g., a distance of object(s) such as the objects 410, 412, 414, 416, etc. to a point associated with the occluded region 418), a trajectory distance to the occlusion (e.g., a distance between a point associated with the trajectory 420 and the occluded region 418), other object(s) information (e.g., a classification, position, velocity, acceleration, pedestrian gaze, traffic light status, and the like of any objects in the environment), a vehicle velocity (e.g., of the vehicle 408), a drivable area (e.g., lane information such as widths, speed limits, road surface type, etc.), map data (e.g., indicative of the objects 414, 416, objects (e.g., buildings, trees, etc.) in the environment), location information (e.g., a location of the environment, a global position, etc.), time (e.g., a time of day, day of the week, season, etc.), and occlusion occupancy (e.g., ground truth information indicating the pedestrian 422 is present but not visible in the occluded region 418 at the time T1), and the like.”; teaches obtaining location information of objects that are occluded, such as the crosswalk and pedestrian 422); and continuing to track the second object after the first object moves away not to occlude the second object (Kobilarov, FIG. 4; col. 11 line 8-21, “The example 406 illustrates the vehicle 408 traversing along the trajectory 420 to a location where the pedestrian 422 is visible to sensors associated with the vehicle 408. In some examples, the data associated with the example 406 can be used as ground truth data to indicate the known occupancy of the occluded region 418 at the time T1. Further, as the vehicle 408 is at a different location a time TN compared to the location of the vehicle at time T1, an occluded region 424 may be associated with the occluding object 410 at the time TN. As noted above, attributes associated with the examples 404 and/or 406 can be used to train a machine learning model to output prediction probabilities associated with an occupancy of an occluded region.”; teaches pedestrian 422 now being visible and the location (one of the attributes) found at TN ). It would have been obvious to a person having ordinary skill in the art before the time of the effective filing date of the claimed invention of the instant application to implement the object attribute identification techniques as taught by Kobilarov with the occlusion region prediction tracking system of Liu in view of Silva. The suggestion/motivation for doing so would have been to “improve a functioning of a computing device, such as a computing device of an autonomous vehicle, in a number of additional ways. In some examples, determining probabilit(ies) that an occluded region is occupied by an object can allow other vehicles or objects to better plan trajectories that ensure safe and comfortable movement through an environment. For example, if a probability associated with an occluded region indicates a high likelihood that the occluded region is occupied by a pedestrian, a vehicle controller may slow the vehicle and/or plan a trajectory to increase a distance or buffer between a planned motion of the vehicle through the environment and the occluded region” (Kobilarov, col. 4 lines 34-46). Further, one skilled in the art could have combined the elements as described above by known methods with no change in their respective functions, and the combination would have yielded nothing more than predictable results. Therefore, it would have been obvious to combine Liu in view of Silva with Kobilarov to obtain the invention as specified in claim 22. Regarding claim 25, arguments made in rejecting claims 22 are analogous. Claims 23 and 26 are rejected under 35 U.S.C. 103 as being unpatentable over Liu in view of Silva and Kobilarov as applied to claims 22 and 25 above, and further in view of Janzen et al. (US 20180096595 A1), hereinafter referred to as Janzen. Regarding claim 23, Liu in view of Silva and Kobilarov teaches the system of claim 22. Liu in view of Silva and Kobilarov fails to teach wherein the predicting the second object location uses a Kalman filter. Janzen teaches wherein the predicting the second object location uses a Kalman filter (Janzen, ¶ 88, “There are cases where an object is no longer moving but was moving in the past and is still in the intersection. The most common case of this is during times of peak traffic where multiple cars may be stuck in an intersection. Since the most computationally efficient algorithm for detecting vehicles is often to look for motion, it is important to track objects that stop moving even after they lose their motion signature. By tracking each object's location, velocity, and acceleration, this can readily be achieved. The preferred embodiment of this prediction and tracking includes a particle filter, such as (but not limited to) a Kalman Filter, potentially using the Hungarian algorithm or another cost-minimization algorithm to match predicted tracks to recent detections”; teaches a Kalman filter that is capable of predicting and tracking an object’s location, velocity, and acceleration which is meant for determining the state of an object when it is unable to be directly observed). It would have been obvious to a person having ordinary skill in the art before the time of the effective filing date of the claimed invention of the instant application to implement the Kalman filter as taught by Janzen onto the device and coordinate object tracking system of Liu in view of Silva and Kobilarov. The suggestion/motivation for doing so would have been to improve object tracking in the case of the measurements of the sensors being noisy or inaccurate. Further, one skilled in the art could have combined the elements as described above by known methods with no change in their respective functions, and the combination would have yielded nothing more than predictable results. Therefore, it would have been obvious to combine Liu in view of Silva and with Janzen to obtain the invention as specified in claim 22. Regarding claim 26, arguments made in rejecting claim 23 are analogous. Claim 5 is rejected under 35 U.S.C. 103 as being unpatentable over Liu in view of Silva as applied to claims 1, 7-12, 14-18, 20, 21, and 24 above, and further in view of Chakraborty et al. (US20200193628A1), hereinafter referred to as Chakraborty. Regarding claim 5, Liu in view of Silva teaches the system of claim 1. Liu in view of Silva fails to teach wherein the real-world coordinates comprise a point on a ground plane of the object in a first image sensor centered real-world coordinate system. Chakraborty teaches wherein the real-world coordinates comprise a point on a ground plane of the object in a first image sensor centered real-world coordinate system (Chakraborty, ¶ 56, “In order to establish targets for regression, 2D box coordinates (x, y) need to be projected to locations (x, yw, zw) on the ground plane. While exact recovery of 3D from 2D points is ill-posed, two reasonable assumptions aid in making the solution tractable: (a) pedestrians rest on ground plane, i.e., yw=0; and (b) the direction of ground plane is parallel to an upright person, i.e., Ĝ=(0, 1, 0). Based on these constraints, the position of an Rd on the ground plane (denoted by G(b)yw=0) is computed algebraically as follows: (a) find the camera center, given by the null space of the camera matrix P=null (P); (b) the ray emanating from the camera center to the object is given by R=inv (P)*Xim, where Xim is the bottom coordinate of an image box; and (c) the intersection between the projected ray R and the ground unit vector Ġ is computed algebraically to give the corresponding touch point Xg on the ground.”; teaches finding a point on the ground plane where a pedestrian is in a coordinate system based on a camera center). It would have been obvious to a person having ordinary skill in the art before the time of the effective filing date of the claimed invention of the instant application to implement the ground plane coordinate system as taught by Chakraborty with the device and coordinate object tracking system of Liu in view of Silva. The suggestion/motivation for doing so would have been that “[b]y inducing geometric context into feature learning, occluded parts may be better modeled” (Chakraborty, ¶31). “Geometric loss is introduced for bounding box regression, where the scene layout is used as a projection space to impose relational and position consistency of proposals on the ground plane. By utilizing the underlying scene geometry to modulate this type of feature activation, discriminative features can be learned that improve the baseline object detector” (Chakraborty, ¶ 30). Further, one skilled in the art could have combined the elements as described above by known methods with no change in their respective functions, and the combination would have yielded nothing more than predictable results. Therefore, it would have been obvious to combine Liu in view of Silva with Chakraborty to obtain the invention as specified in claim 5. 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 Anna Lei whose telephone number is (703)756-5602. The examiner can normally be reached Monday - Friday: 7:30AM - 5:00PM ET. 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 at (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 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. /ANNA LEI/Examiner, Art Unit 2672 /GANDHI THIRUGNANAM/Primary Examiner, Art Unit 2672
Read full office action

Prosecution Timeline

Show 3 earlier events
Nov 18, 2025
Final Rejection mailed — §103
Feb 12, 2026
Response after Non-Final Action
Apr 20, 2026
Request for Continued Examination
Apr 22, 2026
Interview Requested
Apr 23, 2026
Response after Non-Final Action
Apr 28, 2026
Applicant Interview (Telephonic)
May 04, 2026
Examiner Interview Summary
Nov 02, 2026
Non-Final Rejection mailed — §103 (current)

Precedent Cases

Applications granted by this same examiner with similar technology

Patent 12776123
COMPUTER-IMPLEMENTED SYSTEMS AND METHODS FOR INTELLIGENT IMAGE ANALYSIS USING SPATIO-TEMPORAL INFORMATION
3y 3m to grant Granted Oct 27, 2026
Patent 12705725
SYSTEMS AND METHODS FOR PREDICTING A QUALITY OF A PRINTED CIRCUIT BOARD ASSEMBLY
3y 12m to grant Granted Aug 11, 2026
Patent 12705780
LEARNING APPARATUS, LEARNING METHOD, IMAGING APPARATUS, SIGNAL PROCESSING APPARATUS, AND SIGNAL PROCESSING METHOD
3y 2m to grant Granted Aug 11, 2026
Patent 12694499
IMAGING DEVICE AND SYSTEM FOR INSPECTING CABLES AND CABLE JOINTS
4y 3m to grant Granted Jul 28, 2026
Patent 12675701
METHOD FOR FEW-SHOT UNSUPERVISED IMAGE-TO-IMAGE TRANSLATION
7y 5m to grant Granted Jul 07, 2026
Study what changed to get past this examiner. Based on 5 most recent grants.

Strategy Recommendation AI-generated — please review before filing

Get a prosecution strategy drawn from examiner precedents, rejection analysis, and claim mapping.
Typically takes 5-10 seconds — AI-generated, attorney review required before filing

Prosecution Projections

3-4
Expected OA Rounds
78%
Grant Probability
99%
With Interview (+25.1%)
3y 9m (~0m remaining)
Median Time to Grant
High
PTA Risk
Based on 398 resolved cases by this examiner. Grant probability derived from career allowance rate.

Sign in with your work email

Enter your email to receive a magic link. No password needed.

Personal email addresses (Gmail, Yahoo, etc.) are not accepted.

Free tier: 3 strategy analyses per month