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
The following is a quotation of 35 U.S.C. 112(f):
(f) Element in Claim for a Combination. – An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The following is a quotation of pre-AIA 35 U.S.C. 112, sixth paragraph:
An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The claims in this application are given their broadest reasonable interpretation using the plain meaning of the claim language in light of the specification as it would be understood by one of ordinary skill in the art. The broadest reasonable interpretation of a claim element (also commonly referred to as a claim limitation) is limited by the description in the specification when 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is invoked.
As explained in MPEP § 2181, subsection I, claim limitations that meet the following three-prong test will be interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph:
(A) the claim limitation uses the term “means” or “step” or a term used as a substitute for “means” that is a generic placeholder (also called a nonce term or a non-structural term having no specific structural meaning) for performing the claimed function;
(B) the term “means” or “step” or the generic placeholder is modified by functional language, typically, but not always linked by the transition word “for” (e.g., “means for”) or another linking word or phrase, such as “configured to” or “so that”; and
(C) the term “means” or “step” or the generic placeholder is not modified by sufficient structure, material, or acts for performing the claimed function.
Use of the word “means” (or “step”) in a claim with functional language creates a rebuttable presumption that the claim limitation is to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites sufficient structure, material, or acts to entirely perform the recited function.
Absence of the word “means” (or “step”) in a claim creates a rebuttable presumption that the claim limitation is not to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is not interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites function without reciting sufficient structure, material or acts to entirely perform the recited function.
Claim limitations in this application that use the word “means” (or “step”) are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action. Conversely, claim limitations in this application that do not use the word “means” (or “step”) are not being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action.
This application includes one or more claim limitations that do not use the word “means,” but are nonetheless being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, because the claim limitation(s) uses a generic placeholder that is coupled with functional language without reciting sufficient structure to perform the recited function and the generic placeholder is not preceded by a structural modifier. Such claim limitation(s) is/are: “a part detector to determine estimates of second positions of a plurality of joints” in claims 26, 30-31, and 35.
Because this/these claim limitation(s) is/are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, it/they is/are being interpreted to cover the corresponding structure described in the specification as performing the claimed function, and equivalents thereof. In this case, “a part detector to determine estimates of second positions of a plurality of joints” in claims 26, 30-31, and 35 corresponds to a convolutional neural network [Specification, pg. 9].
If applicant does not intend to have this/these limitation(s) interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, applicant may: (1) amend the claim limitation(s) to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph (e.g., by reciting sufficient structure to perform the claimed function); or (2) present a sufficient showing that the claim limitation(s) recite(s) sufficient structure to perform the claimed function so as to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph.
Claim Rejections - 35 USC § 102
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 the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claim(s) 26-28, 30, 35-38, 40, and 44 is/are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Katabi et al. (US 2019/0188533 A1 “KATABI”).
Regarding claim 26, KATABI discloses a method comprising:
using a radar system to transmit an electromagnetic signal, detect reflections of the electromagnetic signal from reflectors, and determine a plurality of data points corresponding with first positions of the reflectors (the sensor subsystem 101 includes a radio 107 connected to a transmit antenna 109 and two receive antenna arrays 108 and 110. The radio is configured to transmit a low power RF signal into the environment 103 using the transmit antenna 109. Reflections of the transmitted signal are received at the radio 107 through the receive antenna arrays 108, 110 [0030-0031])
and processing the data points to determine a posture of a subject by:
using a part detector to determine estimates of second positions of a plurality of joints (the sequences of two-dimensional RF heatmaps 112, 114 are processed by the keypoint estimation module to generate a sequence of estimated keypoint confidence maps 118 indicating an estimated location of keypoints (e.g., legs, arms, hands, feet, etc.) of a subject (e.g., a human body) in the environment 103 [0028])
and using a spatial model to refine the estimates (the training process results in a keypoint estimation module 102 that accounts for the properties of RF signals such as specularity of the human body, low spatial resolution, and invariance to translations in both space and time [0061])
wherein the spatial model encodes expected relative positions between the joints (the sequence of estimated keypoint confidence maps 118 generated by the keypoint estimation module 102 is provided to a keypoint association module 124 which maps the keypoints in the estimated confidence maps 118 to depictions of posed skeletons 134 [0052]); (the pose estimation CNN learns to infer the localization of occluded keypoints based on the locations of other keypoints [0086]).
Regarding claim 27, KATABI discloses the method of claim 26, wherein the part detector comprises a convolutional neural network that has been trained to determine the estimates (the pose estimation CNN learns to infer the localization of occluded keypoints based on the locations of other keypoints [0086], cited and incorporated in the rejection of claim 26).
Regarding claim 28, KATABI discloses the method of claim 26, further comprising performing a temporal correlation operation to smooth an output from the spatial model (spatiotemporal convolutions are used as basic building blocks for the keypoint estimation module 102 [0048]).
Regarding claim 30, KATABI discloses the method of claim 26, further comprising: a two-dimensional (2D) image from the data points; and providing the 2D image as an input to the part detector (the sequences of two-dimensional RF heatmaps 112, 114 are processed by the keypoint estimation module to generate a sequence of estimated keypoint confidence maps 118 indicating an estimated location of keypoints (e.g., legs, arms, hands, feet, etc.) of a subject (e.g., a human body) in the environment 103 [0028], cited and incorporated in the rejection of claim 26).
Regarding step 35, KATABI discloses a system for determining the posture of a subject using a radar system, comprising:
a radar system configured to: transmit an electromagnetic signal; detect reflections of the electromagnetic signal from reflectors; and determine a plurality of data points corresponding with first positions of the reflectors (the sensor subsystem 101 includes a radio 107 connected to a transmit antenna 109 and two receive antenna arrays 108 and 110. The radio is configured to transmit a low power RF signal into the environment 103 using the transmit antenna 109. Reflections of the transmitted signal are received at the radio 107 through the receive antenna arrays 108, 110 [0030-0031])
and a processor configured to process the data points to determine a posture of a subject by:
using a part detector to determine estimates of second positions of a plurality of joints (the sequences of two-dimensional RF heatmaps 112, 114 are processed by the keypoint estimation module to generate a sequence of estimated keypoint confidence maps 118 indicating an estimated location of keypoints (e.g., legs, arms, hands, feet, etc.) of a subject (e.g., a human body) in the environment 103 [0028])
and using a spatial model to refine the estimates (the training process results in a keypoint estimation module 102 that accounts for the properties of RF signals such as specularity of the human body, low spatial resolution, and invariance to translations in both space and time [0061])
wherein the spatial model encodes expected relative positions between joints (the sequence of estimated keypoint confidence maps 118 generated by the keypoint estimation module 102 is provided to a keypoint association module 124 which maps the keypoints in the estimated confidence maps 118 to depictions of posed skeletons 134 [0052]); (the pose estimation CNN learns to infer the localization of occluded keypoints based on the locations of other keypoints [0086]).
Regarding claim 36, KATABI discloses the system of claim 35, further comprising a plurality of radar systems configured to provide data points to the processor, wherein the processor is further configured to use the data points from the radar system to determine the estimates (the sensor subsystem 101 includes a radio 107 connected to a transmit antenna 109 and two receive antenna arrays 108 and 110. The radio is configured to transmit a low power RF signal into the environment 103 using the transmit antenna 109. Reflections of the transmitted signal are received at the radio 107 through the receive antenna arrays 108, 110 [0030-0031], cited and incorporated in the rejection of claim 26).
Claims 37-38 correspond to respective claims 27-28 sufficiently in scope and therefore are similarly rejected.
Claim 40 corresponds to respective claim 30 sufficiently in scope and therefore is similarly rejected.
Regarding claim 44, KATABI discloses the method of claim 26, wherein the joints comprise a head, a left shoulder, a right shoulder, a left hip, a right hip, a left elbow, a right elbow, a left knee, and a right knee (the pose estimation problem is defined as generating two-dimensional (i.e., 2-D) or three-dimensional (i.e., 3-D) skeletal representations of the joints on the arms and legs, and keypoints on the torso and head [0004]), wherein the spatial model defines primary joints as the head, the left shoulder, the right shoulder, the left hip, and the right hip, defines secondary joints as the left elbow, the right elbow, the left knee, and the right knee (RF signals in the environment are used to extract full three-dimensional (i.e., 3-D) poses/skeletons of multiple subjects (including the head, arms, shoulders, hip, legs, etc.) [0009]), defines the expected relative position of each of the primary joints as dependent on the second positions of the other primary joints, and defines the expected relative position of each of the secondary joints as dependent on the second positions of the primary joints (the pose estimation CNN learns to infer the localization of occluded keypoints based on the locations of other keypoints [0086], cited and incorporated in the rejection of claim 26).
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(s) 29 and 39 is/are rejected under 35 U.S.C. 103 as being unpatentable over KATABI, in view of Craig (US 2013/0156260 A1 “CRAIG”).
Regarding claim 29, KATABI discloses the method of claim 28, wherein the temporal correlation operation comprises determining, for each of the estimates, a confidence level (for each pixel of a given RGB frame 116 in the sequence of RGB frames 116, the corresponding keypoint confidence map 118 indicates the confidence that the pixel is associated with a particular keypoint (e.g., the confidence that the pixel is associated with a hand or a head) [0056]). However, KATABI does not disclose determining a speed of movement, and wherein the temporal correlation operation rejects updated joint positions in response to the confidence level and/or the speed of movement.
In a same or similar field of endeavor, CRAIG teaches that examples of dimensions include a confidence or probability in a position of a skeletal point, a velocity and/or acceleration vector for some or all of the skeletal points [0126]. An indication of zero, one, or more problem states (e.g., problem state 103) may be output from the background removal process 104 [0069]. Example problem states may include: a velocity limited state in which a portion of the human subject moves at a rate that exceeds an upper or lower velocity threshold [0018]. It is further noted that the limitation is in alternative form; therefore, only one alternative was given patentable weight.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of KATABI to include the teachings of CRAIG, because doing so would improve accuracy and certainty of the pose recognition system, as recognized by CRAIG.
Claim 39 corresponds to respective claim 29 sufficiently in scope and therefore is similarly rejected.
Claim(s) 31 and 32 is/are rejected under 35 U.S.C. 103 as being unpatentable over KATABI, in view of Moawad et al. (US 2022/0197301 A1 “MOAWAD”).
Regarding claim 31, KATABI discloses (Examiner’s note: What KATABI does not clearly disclose is ) a method comprising:
obtaining ground truth positions of joints (the keypoint confidence maps 118′ generated by the teacher network 104 are treated as ground truth [0056]) while the sensor subsystem 101 includes a radio 107 connected to a transmit antenna 109 and two receive antenna arrays 108 and 110. The radio is configured to transmit a low power RF signal into the environment 103 using the transmit antenna 109. Reflections of the transmitted signal are received at the radio 107 through the receive antenna arrays 108, 110 [0030-0031])
training a part detector to determine estimates of second positions of the joints from data points corresponding to first positions of reflectors by minimizing a first loss function determined with reference to the ground truth positions (the training objective of the keypoint estimation module S(⋅) is to minimize the difference between its estimation S(R) and the teacher network's estimation T(I) [0059]); (the keypoint confidence maps 118′ generated by the teacher network 104 are treated as ground truth [0056]); (to localize a keypoint, the CNN outputs scores s={sν}ν∈V corresponding to all 3-D voxels ν∈V, and the target voxel ν* is the one that contains the keypoint. SoftMax loss LSoftmax(s,ν*) is used as the looks for keypoint localization [0073])
and training the part detector to refine the estimates by minimizing a second loss function determined with reference to the ground truth positions (the model to localize all of the keypoints jointly and infers the localization of occluded keypoints based on the locations of other keypoints. The total loss of pose estimation is the sum of the SoftMax loss of all 14 keypoints [0074])
However, KATABI does not explicitly disclose obtaining positions while concurrently detecting reflections.
In a same or similar field of endeavor, MOAWAD teaches a radar-localization module that may be used to implement vehicle localization based on radar detections. In the example illustration 200-3, the radar-localization module 210 is configured to be in a real-time localization mode. The output of the radar-localization module 210 in real-time localization mode is an updated vehicle pose 236 of the vehicle 104 [0047]. In real-time localization mode, the scan-matcher 224 receives the NDT radar reference map 234 as input, in addition to the landmark data and the ego-trajectory information. The inputs are used by the scan-matcher to determine an NDT grid. The NDT grid is compared to the NDT radar reference map to determine the updated vehicle pose 236 [0048].
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of KATABI to include the teachings of MOAWAD, because doing so would enable real-time, accurate and dynamic target detection, as recognized by MOAWAD.
Regarding claim 32, KATABI/ MOAWAD discloses the method of claim 31, further comprising performing step ii)) and step iii) sequentially (the model to localize all of the keypoints jointly and infers the localization of occluded keypoints based on the locations of other keypoints. The total loss of pose estimation is the sum of the SoftMax loss of all 14 keypoints [KATABI 0074]). The Examiner further noted that the sequential order is implied in KATABI where each keypoint SoftMax loss is determined and then the total SoftMax loss is determined.
Claim(s) 33 is/are rejected under 35 U.S.C. 103 as being unpatentable over KATABI, in view of MOAWAD, and further in view of Mathew et al. (US 2021/0279550 A1 “MATHEW”).
Regarding claim 33, KATABI/ MOAWAD discloses the method of claim 31. However, KATABI/ MOAWAD does not disclose wherein the training in step i) and/or step ii) comprises performing a gradient descent method.
In a same or similar field of endeavor, MATHEW teaches that the coefficients or weights of the filters are determined by training the CNN with a set of training images. FIG. 2 is a simplified example illustrating training of a CNN using a commonly used training technique referred to as stochastic gradient descent back propagation [0031]. The speed of adjustment of coefficient values is controlled by the “learning rate” which is a multiplication factor with a small value that is applied to the gradients during back propagation. This fine tuning is modified to include a sparsification technique. The sparsified fine tuning is explained in more detail in reference to FIGS. 12-14 [0052].
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of KATABI to include the teachings of MATHEW, because gradient descent is used to update all the filter coefficients to minimize the total error, as recognized by MATHEW.
Claim(s) 34 is/are rejected under 35 U.S.C. 103 as being unpatentable over KATABI, in view of MOAWAD, and further in view of Hu et al. (US 2020/0265567 A1 “HU”).
Regarding step 34, KATABI/ MOAWAD discloses the method of claim 31. However, KATABI/ MOAWAD does not disclose wherein the training in steps i) and ii) uses a dynamic learning rate of between 10-2 and 10-5.
In a same or similar field of endeavor, HU teaches that the overall goal of the optimization is to try and reduce or minimize the loss function. In some embodiments, the “Adam” solver (which is derived from adaptive moment estimation) known in the art could be used to optimize the convolutional neural network's weights with a learning rate of 0.01. However, other optimization techniques and hyper-parameters (such as the learning rate) for the convolutional neural network could be used here [0094].
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of KATABI to include the teachings of HU, because doing so would optimize the CNN training for target detection system, as recognized by HU.
Claim(s) 41 and 45 is/are rejected under 35 U.S.C. 103 as being unpatentable over KATABI, in view of Pasupuleti et al. (US 2021/0282667 A1 “PASUPULETI”).
Regarding claim 41, KATABI discloses the system of claim 35. However, KATABI does not explicitly disclose that wherein the radar system comprises a millimeter-wave (mmWave) radar system, and wherein the electromagnetic signal has a frequency of between 75 gigahertz (GHz) and 85 GHz.
In a same or similar field of endeavor, PASUPULETI teaches receiving, in a first input layer of the MLNN, from a millimeter wave (mmWave) radar sensing device, a first set of mmWave radar point cloud data representing a first gait characteristic of the subject [0021]. Millimeter wave radar sensing technology as described and applied herein refers to detection of objects and providing information on range, velocity and angle of those objects. mmWave radar uses a contactless technology which operates in the spectrum between 30 GHz and 300 GHz [0008].
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of KATABI to include the teachings of PASUPULETI, because utilizing mmWave is simple substitution of known system to obtain predictable results of determining targets.
Regarding claim 45, KATABI discloses the method of claim 26. However, KATABI does not disclose that wherein the electromagnetic signal has a frequency of between 75 gigahertz (GHz) and 85 GHz.
In a same or similar field of endeavor, PASUPULETI teaches receiving, in a first input layer of the MLNN, from a millimeter wave (mmWave) radar sensing device, a first set of mmWave radar point cloud data representing a first gait characteristic of the subject [0021]. Millimeter wave radar sensing technology as described and applied herein refers to detection of objects and providing information on range, velocity and angle of those objects. mmWave radar uses a contactless technology which operates in the spectrum between 30 GHz and 300 GHz [0008].
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of KATABI to include the teachings of PASUPULETI, because utilizing mmWave is simple substitution of known system to obtain predictable results of determining targets.
Claim(s) 42 is/are rejected under 35 U.S.C. 103 as being unpatentable over KATABI, in view of Stadelmann et al. (US 9,568,602 B1 “STADELMANN”).
Regarding claim 42, KATABI discloses the system of claim 36, wherein the radar systems comprise a first radar system with a first boresight and a second radar system with a second boresight (the sensor subsystem 101 includes a radio 107 connected to a transmit antenna 109 and two receive antenna arrays: a vertical antenna array 108 and a horizontal antenna array 110 [0030]). However, KATABI does not disclose that wherein the first boresight is at an angle of at least 25 degrees (°) relative to the second boresight.
In a same or similar field of endeavor, STADELMANN teaches that with reference to FIG. 6, the radar antenna 104 includes a first panel 602 and a second panel 602. The first and second panels 602 and 606 are arranged side-by-side and adjacent with respect to each other. In some embodiments, the first and second panels 602 and 606 are placed in a V shape defining an angle of approximately 40-60 degrees [col. 10, lines 19-25]. It is further noted that it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify KATABI to include relative boresight of at least 25 degrees (°), since it has been held that discovering an optimum value of a result effective variable involves only routine skill in the art. In re Boesch, 617 F.2d 272, 205 USPQ 215 (CCPA 1980).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of KATABI to include the teachings of STADELMANN, because doing so would improve target detection while minimizing system size and cost, as recognized by STADELMANN.
Claim(s) 43 is/are rejected under 35 U.S.C. 103 as being unpatentable over KATABI, in view of STADELMANN, and further in view of Baker et al. (US 2021/0298643 A1 “BAKER”).
Regarding claim 43, KATABI/ STADELMANN discloses the system of claim 42, wherein the first radar system is separated from the second radar system (the sensor subsystem 101 includes a radio 107 connected to a transmit antenna 109 and two receive antenna arrays: a vertical antenna array 108 and a horizontal antenna array 110 [0030]). However, KATABI/ STADELMANN does not explicitly disclose that the separation is by a distance of at least 0.5 meters (m), and wherein the first radar system and the second radar system are at a height of between 1 m and 2 m.
In a same or similar field of endeavor, BAKER teaches a system 200 for monitoring a patient using radar sensors includes the radar sensors 106, 108, 110 [0216]. Furthermore, a radar sensor 3402 is mounted on a mobile physical therapy device 3406. Mobile physical therapy device 3406 further includes a generally vertically oriented pole or mast 3412 extending upwardly from base 3408. A pivotable arm 3414 extends from an upper region 3416 of pole 3412 and radar sensor 3402 is mounted to a distal end of arm 3414 in spaced relation with pole 3412. Arm 3414 is pivotable upwardly and downwardly relative to pole 3412 to adjust a height at which radar sensor 3402 is supported above the floor [0304]. It is further noted that although BAKER does not explicitly teach separation distance of at least 0.5 m and radar system height of between 1 m and 2 m, BAKER teaches that radar sensors may be installed in various locations and are adjustable. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify KATABI to include separation distance of at least 0.5 m and radar system height of between 1 m and 2 m, since it has been held that discovering an optimum value of a result effective variable involves only routine skill in the art. In re Boesch, 617 F.2d 272, 205 USPQ 215 (CCPA 1980).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of KATABI to include the teachings of BAKER, because doing so would provide suitable and mobile solutions for intensive, prolonged care for patient monitoring while minimizing cost, as recognized by BAKER.
Claim(s) 46 is/are rejected under 35 U.S.C. 103 as being unpatentable over KATABI, in view of Cao et al. (CN 111353447 A “CAO”).
Regarding claim 46, KATABI discloses the method of claim 26. However, KATABI does not disclose wherein the spatial model is based on a dependency graph implemented as a convolution kernel, and wherein the method further comprises initializing kernel weights of kernels of the convolutional kernel by selecting pair-wise position dependency between the joints from ground truth positions.
In a same or similar field of endeavor, CAO teaches identifying two-person interactive behavior based on a graph convolutional network. The identification method includes constructing an intrinsic dependency graph of human joints, an individual extrinsic dependency graph, and an interaction dependency graph; assigning different weights to the connection edges of the three joint connection graphs; sending them into a graph convolutional network for learning and extracting spatial features; sending the spatial features obtained based on each frame into a long-short term memory network for temporal modeling; and obtaining the recognition results of the interactive behavior categories [0008].
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of KATABI to include the teachings of CAO, because doing so would improve target recognition performance, as recognized by CAO.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to HAILEY R LE whose telephone number is (571)272-4910. The examiner can normally be reached 9:00 AM - 5:00 PM EST.
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 at (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.
/Hailey R Le/Examiner, Art Unit 3648 September 10, 2025
/William Kelleher/Supervisory Patent Examiner, Art Unit 3648