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 .
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.
Claim Rejections - 35 USC § 103
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 of this title, 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.
The factual inquiries set forth in Graham v. John Deere Co., 383 U.S. 1, 148 USPQ 459 (1966), that are applied for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1 – 4, and 8 – 11 are rejected under 35 U.S.C. 103 as being unpatentable over Fujimura et al. (Publication: US 2008/0152191 A1) in view of Tian et al. (Publication: US 2020/0126297 A1).
Regarding claim 1, Fujimura discloses a system for improving the accuracy of modeling human-object interaction tracking, the system comprising ([0029], [0032] – The system includes the RAM stores instructions to be performed by the processor to implemented the following methods including tracking a pose of a subject.):
a processor configured to ([0029], [0032] – The system includes the RAM stores instructions to be performed by the processor configured to perform:):
receive, from a camera, first data corresponding to a first image and a second image of a human and an object (
[0040] FIG. 2 illustrates a flowchart of a method of estimating and tracking the poses of the human body 100.
First, the pose of the subject human body 100 is initialized 210 using a pose (e.g., T-pose where the subject faces the depth camera 110 with both arms stretching down and spread sideways). Receive the next sequences of images and pose from the video camera 112 and the depth camera and then the human body 100 and pose, “first image and a second image of a human and an object”. );
process the first data to generate a mesh for the human and the object ( [0044] - Fig. 4 - Pose Vector is generated using previous frame 410.
[0062] - Fig. 8 – Obtain vertices of model using previous frame, process the previous frame to generate the vertices and vector for the human body and pose.
It is known that vertices are the fundamental building blocks of a 3D mesh in computer graphics.);
obtain pose data by sampling the pose distribution ([0071] confirm the robustness of the embodiment for estimating and tracking poses involving fast movements. Sampling interval Tracking error for Left Hand. That is, only one image from k consecutive images was taken and Table 3 “sampling the pose distribution” tracks the error for the post distribution in consecutive images in different time.
PNG
media_image1.png
164
364
media_image1.png
Greyscale
);
receive second data corresponding to the second image and a third image of the human and the object (
[0044] - Fig. 4 - Labeling Module receives the current depth image 132 and the current color images that are related to human and its pose. ) ; and
process the second data and the sampled pose data to generate an updated mesh for the human and the object (
[0044], [0048], [0071] - Fig. 4, current depth image 132, the current color images and Labeling module 416 includes optimization engine 612 on tracking sampling interval tracking error, Table 3 to generate updated pose vector of the estimates human body and the pose
[0038] - The pose estimator 118 generates pose vector 130 of the estimated pose of the human body 100.
It is known that mesh is fundamentally build from vectors, collection of the point of vertices.).
Fujimura does not disclose; however Tian discloses
generate a pose distribution using the mesh ([0080] - the pose θ articulates how the 3D surface of the human body mesh deforms, generate, in accordance with the distribution of the physical key points and different poses of the corresponding physical key points.)
Before the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art to modify Fujimura with generate a pose distribution using the mesh as taught by Tian. The motivation for doing is to accurately calculate the human body in order to provide an effective and efficient human-machine interface as taught by Tian.
Regarding claim 2, Fujimura in view of Tian disclose all the limitation of claim 1.
Fujimura discloses receiving motion data for the human and the object from one or more motion sensors ([0069] - A motion capture system was coupled with eight cameras to obtain actual coordinates of eight major joints of the subject human for comparison with the estimated coordinates of the joints obtained using the embodiment. In the experiments, markers were attached to the subject human to generate the actual coordinates.); and
optimizing the sampling using the motion data ([0053] The optimization engine 612 groups the segments s.sub.i (i=1,2, . . . ,N) obtained from the segment generator 610 into the labeled parts {p.sub.1,p.sub.2, . . . ,p.sub.M}. In the equation, c(i, j) represents the Euclidean distance from a segment si to pixels sampled from the labeled part P.sub.j in the model that is derived from the pose vector 410 of the previous image to solve optimization thus optimization is achieved.
PNG
media_image2.png
252
320
media_image2.png
Greyscale
).
Regarding claim 3, Fujimura in view of Tian disclose all the limitation of claim 1.
Fujimura discloses the first data includes data for the first image and for the second image ([0040] FIG. 2 illustrates a flowchart of a method of estimating and tracking the poses of the human body 100.
First, the pose of the subject human body 100 is initialized 210 using a pose (e.g., T-pose where the subject faces the depth camera 110 with both arms stretching down and spread sideways). Receive the next sequences of images and pose from the video camera 112 and the depth camera and then the human body 100 and pose, “first image, a second image of a human and an object”.);
wherein the first data includes human and object segmentation data for the first image and for the second image (
[0040] FIG. 2 illustrates a flowchart of a method of estimating and tracking the poses of the human body 100.
First, the pose of the subject human body 100 is initialized 210 using a pose (e.g., T-pose where the subject faces the depth camera 110 with both arms stretching down and spread sideways). Receive the next sequences of images and pose from the video camera 112 and the depth camera and then the human body 100 and pose, “first image, a second image of a human and an object”.
[0010] - (i) grouping data points, human and pose, obtained from image capturing devices into group, segmentation. ).
Tian discloses
the data includes RGB data ( [0076] - the captured 2D image is RGB image, can be a full-body image of the human subject 104, an image of an upper body of the human subject 104, an image of a lower body of the human subject 104, or a significant portion of the human body for which posture change is sufficiently discernable.)
Regarding claim 4, Fujimura in view of Tian disclose all the limitation of claim 1.
Tian discloses to process the first data by using at least one neural network ([0078] - the computing system uses a trained model (e.g., a deep neural network model including multiple layers, such as ResNet 50) to obtain the convolutional features of the captured 2D image.) .
Before the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art to modify Fujimura in view Tian with to process the first data by using at least one neural network as taught by Tian. The motivation for doing is to accurately calculate the human body in order to provide an effective and efficient human-machine interface.
Regarding claim 8, see rejection on claim 1.
Regarding claim 9, see rejection on claim 2.
Regarding claim 10, see rejection on claim 3.
Regarding claim 11, see rejection on claim 4.
Claims 5, 6, 12, and 13 are rejected under 35 U.S.C. 103 as being unpatentable over Fujimura et al. (Publication: US 2008/0152191 A1) in view of Tian et al. (Publication: US 2020/0126297 A1) and Seff et al. (Publication: US 2024/0300542 A1).
Regarding claim 5, Fujimura in view of Tian disclose all the limitation of claim 4.
Fujimura in view of Tian do not disclose; however, Seff discloses
wherein the at least one neural network includes a self-attention layer( [0092] The trajectory decoder neural network 406 can generate the sequences of discrete motion tokens 204 by transforming sequences of motion tokens using attention mechanisms. For example, the trajectory decoder neural network 406 can include self-attention layers that apply self-attention over sequences of motion tokens. While generating the sequence of discrete motion tokens 204, the trajectory decoder neural network 406 can, at each time step, process and transform the sequence of preceding motion tokens using the self-attention layers.).
Before the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art to modify Fujimura in view Self with wherein the at least one neural network includes a self-attention layer as taught by Self. The motivation for doing is to have a more complex algorithm to accurate predicate the joint behavior as taught by Seff.
Regarding claim 6, Fujimura in view of Tian disclose all the limitation of claim 4.
Fujimura in view of Tian do not disclose; however, Seff discloses
wherein the at least one neural network includes a cross-attention layer( [0092] - 406 can include cross-attention layers that apply cross-attention with the scene encoding 404 to sequences of motion tokens. While generating the sequence of discrete motion tokens 204, the trajectory decoder neural network 406 can, at each time step, process and transform the sequence of preceding motion tokens conditioned on the respective scene encoding 404 for the time step using the cross-attention layers.).
Before the effective filing date of the claimed invention, it would have been obvious to one of ordinary skill in the art to modify Fujimura in view Self with wherein the at least one neural network includes a cross-attention layer as taught by Self. The motivation for doing is to have a more complex algorithm to accurate predicate the joint behavior as taught by Seff.
Regarding claim 12, see rejection on claim 5.
Regarding claim 13, see rejection on claim 6.
Allowable Subject Matter
Regarding claims 7 and 14 , No art was found that could fully teach the claim as recited and would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
The closest prior art discovered is the combination of Fujimura et al. (Publication: US 2008/0152191 A1) in view of Tian et al. (Publication: US 2020/0126297 A1), Zhao et al. (Publication: US 2024/0265586 A1), and Chen (Publication: US 2024/0256862 A1).
However, none of the prior art cited above, nor any other prior art discovered by Examiner, fully teaches claims 7 and 14, either singly or in an obvious combination.
Regarding claim 15, No art was found that could fully teach the claim as recited.
The closest prior art discovered is the combination of Fujimura et al. (Publication: US 2008/0152191 A1) in view of Tian et al. (Publication: US 2020/0126297 A1), Zhao et al. (Publication: US 2024/0265586 A1), and Vianello et al. (Publication: US 2023/0026278 A1).
However, none of the prior art cited above, nor any other prior art discovered by Examiner, fully teaches claim 15, either singly or in an obvious combination.
Dependent claims not mentioned specifically above inherit the deficiencies from the claims stated above on which they depend.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to MING WU whose telephone number is (571)270-0724. The examiner can normally be reached on Monday - Thursday and alternate Fridays: 9:30am - 6:00pm 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, Devona Faulk can be reached on 571-272-7515. 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.
/MING WU/
Primary Examiner, Art Unit 2618
Prosecution Insights