DETAILED ACTION
Preliminary Remarks
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Priority
This application claims the bene of application no. 63/613,664 filed 12/21/2023
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) 1-3, 8, 10-12, 18 and 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Oreshkin et al. (U.S. Publication 2024/0054671) and Wang et al. (U.S. Publication 2023/0410397).
In reference to claim 1, Oreshkin et al. discloses a computer-implemented method for generating a pose for a virtual character (see paragraphs 2, 19, 20-21, 131 and Figures 1, 3 and 9 wherein Oreshkin et al. discloses a method of estimating a pose for a custom character in the field of animation, the method performed via a computer processing system and is directed to digital environments including virtual reality and augmented reality environments each comprising virtual components.), comprising:
determining a graph representation of one or more sets of joints in the virtual character based on (i) a set of constraints associated with one or more joints included in the one or more sets of joints and (ii) a set of proportions associated with pairs of joints included in the one or more sets of joints (see paragraphs 3, 22, 27-28, 30-33, 134, 138-139, 144-147 and Figures 1, 3 and 9 wherein Oreshkin et al. discloses the system and method executing a learned IK (inverse kinematics) solver for pose estimations. Oreshkin et al. explicitly discloses the learned solver taking the form of a machine-learned algorithm which is trained and then utilizing in a prediction phase. Oreshkin et al. discloses the algorithm specifically as a neural network that is stored in computational graph formats such as the ONNX file format which is comprises a directed graph with nodes. Oreshkin et al. discloses feeding effectors as input to the neural network or machine learned algorithm, to determine pose of a character, the effectors defining many aspects of constraints for joints of a character which in turn configure the pose for the character. Oreshkin et al. explicitly discusses joint effectors such as reach effectors, look-at effectors and rotational effectors. Oreshkin et al. discloses reach effectors to provide input as to the desired target position in a world space of which the Examiner interprets functionally equivalent to Applicant’s “set of constraints” claim element. Oreshkin et al. discloses the rotational effectors including directional data such as vector information relative to a set of joints. Note, the Examiner interprets such input effectors functionally equivalent to Applicant’s “set of proportions” as the definition of such an element further discusses the association with distances to other joints (see for example dependent claim 10).);
generating, via execution of a first neural network, a set of updated node states for the one or more sets of joints based on the graph representation; and
generating, based on the set of updated node states, one or more output poses that correspond to the one or more sets of joints, wherein the one or more output poses include (i) a first set of joint positions for the one or more sets of joints, (ii) a first set of joint orientations for the one or more sets of joints, and (iii) the set of proportions (see paragraphs 42 and #102, 110, 172, 176, 178 of Figure 1 wherein Oreshkin et al. passing the input effectors through the learned IK (inverse kinematics) solver/neural network and outputting global joint position data as well as local joint rotation data for each input joint which together define a full pose representation. Note, since Oreshkin et al. discloses the rotational effectors including directional data such as vector information relative to a set of joints, it is clear that such effectors at least inherently also include “orientations” as well as the above explained “proportions” elements of the claims).
Although Oreshkin et al. does disclose utilizing a machine learning/neural network in the form of a directed graph comprising nodes, as the IK solver, Oreshkin et al. does not explicitly disclose updating the nodes for the joints based on the graph representation. Wang et al. discloses utilizing neural networks for training and inference to produce desirable realistic images or videos of a character performing an action such as walking, jumping etc. (see paragraphs 1-2). Wang et al. discloses utilizing a generative neural network to produce a character graph where each joint of a character is represented by a node (see paragraph 57 and Figure 2). Wang et al. discloses utilizing such character graphs with physics-based motion controllers to create different poses and moveable joints by taking current state information for various nodes and updating nodes of the graph through the neural network (see at least paragraphs 58-60 and Figures 3-4). It would have been obvious to one of ordinary skill in the art at the time of filing of the invention to implement the motion controller neural network graph based processing techniques of Wang et al. with the IK solver techniques of Oreshkin et al. in order to use machine learning models to replace specific, narrow algorithms to allow a single system to handle broader range of use cases without re-coding underlying graphs and/or graph logic.
In reference to claims 2 and 12, Oreshkin et al. and Wang et al. disclose all of the claim limitations as applied to claims 1 and 11 respectively in addition, Oreshkin et al. explicitly discloses implementing losses within the neural network training/learning and specifically details a first L2 loss that is used as a loss type to penalize errors of 3D joint positions based upon a ground truth (see paragraph 61-63). Oreshkin et al. further explicitly discloses a second loss, a geodesic loss that is a loss type to penalize errors in rotational output of joints or orientations based upon a ground truth rotation matrix (see paragraphs 63-64).
In reference to claim 3, Oreshkin et al. and Wang et al. disclose all of the claim limitations as applied to claim 2 above in addition, Oreshkin et al. further discloses an additional loss, a look-at loss that maybe further used as a loss type associated with the look-at effector (see paragraph 66).
In reference to claim 8, Oreshkin et al. and Wang et al. disclose all of the claim limitations as applied to claim 1 above. Oreshkin et al. explicitly discusses joint effectors such as reach effectors, look-at effectors and rotational effectors (see paragraph 28). Oreshkin et al. discloses reach effectors to provide input as to the desired target position in a world space while the look-at effectors providing a desired target look position for the joins (see paragraphs 31-32.
In reference to claim 10, Oreshkin et al. and Wang et al. disclose all of the claim limitations as applied to claim 1 above. Oreshkin et al. discloses the rotational effectors including directional data such as vector information relative to a set of joints (see paragraph 33). Note, the Examiner interprets such input effectors functionally equivalent to Applicant’s “set of proportions”. Oreshkin et al. further discloses determining skeleton features including specifically determining distances between joint pairs (see paragraph 86 and Figure 3).
In reference to claim 11, claim 11 is similar in scope to claim 1 and is therefore rejected under like rationale. In addition to the rationale as applied in the rejection of claim 1 above, claim 11 further recites, “One or more non-transitory computer-readable media storing instructions that, when executed by one or more processors, cause the one or more processors to perform…” Oreshkin et al. discloses the invention being performed via a computer system comprising a computer readable medium with software modules embodied thereon and executed by hardware processor(s) (see paragraphs 102-104). Wang et al. also discloses the invention being performed via a machine-readable medium having stored thereon instructions which are performed by one or more processors (see at least paragraphs 466 and 611).
In reference to claim 18, Oreshkin et al. and Wang et al. disclose all of the claim limitations as applied to claim 11 above. Oreshkin et al. discloses the IK pose determination to further involve an embedding of data that is semantic and that further describes an intended use of an effector for the joints (see at least paragraph 36) of which the Examiner interprets functionally equivalent to Applicant’s “style associated with the virtual character.”
In reference to claim 20, claim 20 is similar in scope to claim 1 and is therefore rejected under like rationale. In addition to the rationale as applied in the rejection of claim 1 above, claim 20 further recites, “A system, comprising: one or more memories that store instructions; and one or more processors that are coupled to the one or more memories and, when executing the instructions, are configured to perform operations…” Oreshkin et al. discloses the invention being performed via a computer system comprising a computer readable medium with software modules embodied thereon and executed by hardware processor(s) (see paragraphs 102-104). Wang et al. also discloses the invention being performed via a machine-readable medium having stored thereon instructions which are performed by one or more processors (see at least paragraphs 466 and 611). Further, claim 20 requires the graph representation of the one or more joints be based additionally on “(iv) a style associated with the virtual character.” Oreshkin et al. discloses the IK pose determination to further involve an embedding of data that is semantic and that further describes an intended use of an effector for the joints (see at least paragraph 36) of which the Examiner interprets functionally equivalent to Applicant’s “style associated with the virtual character.”
Allowable Subject Matter
Claims 4-7, 9, 13-17 and 19 are objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
References Cited
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
Zhang et al. (U.S. Patent 11,275,931)
Zhang et al. discloses a human pose prediction method that uses a basic neural network to perform prediction on an input target image.
Cashman et al. (U.S. Publication 2023/0326135)
Cashman et al. discloses a method for virtually representing human body poses using a previously-trained pose optimization model.
Ali Akbarian et al. (U.S. Publication 2024/0265659)
Ali Akbarian et al. discloses a method of updating a pose of a plurality of joints of a kinematic tree of an articulated object.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Antonio Caschera whose telephone number is (571) 272-7781. The examiner can normally be reached Monday-Friday between 6:30 AM and 2:30 PM EST.
If attempts to reach the examiner by telephone are unsuccessful, the examiner's supervisor, Said Broome, can be reached at (571) 272-2931.
Any response to this action should be mailed to:
Mail Stop ____________
Commissioner for Patents
P.O. Box 1450
Alexandria, VA 22313-1450
or faxed to:
571-273-8300 (Central Fax)
See the listing of “Mail Stops” at http://www.uspto.gov/patents/mail.jsp and include the appropriate designation in the address above.
Any inquiry of a general nature or relating to the status of this application or proceeding should be directed to the Technology Center 2600 Customer Service Office whose telephone number is (571) 272-2600.
/Antonio A Caschera/
Primary Examiner, Art Unit 2612
6/23/26
Prosecution Insights