CTNF 18/830,218 CTNF 88481 DETAILED ACTION Notice of Pre-AIA or AIA Status 07-03-aia AIA 15-10-aia The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA. Claim Rejections - 35 USC § 102 07-07-aia AIA 07-07 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 – 07-12-aia AIA (a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention. 07-15-03-aia AIA Claim s 1, 2, 4-6, 8-12, 14, and 17-20 are rejected under 35 U.S.C. 102(a)(2) as being anticipated by Starke et al. (US 12,138,543; hereinafter “Starke”) . Regarding claim 1 , Starke discloses A computer-implemented method for animating characters (“animation generation,” abstract) , the method comprising: receiving a first state of a character (Character Control Variables 116 of Fig. 1B; “the character control variables 116 … reflect positions, orientations, and so on, of bones, end effectors, and so on, of a skeleton or rig (e.g., character pose),” col. 8, line 65 – col. 9, line 15) and one or more constraints on one or more motions associated with a subset of joints belonging to the character (User Input 114 of Fig. 1B; “the end user may provide user input 114 to maintain the running … the user input 114 may also indicate that the in-game character is to cease running or perform another movement,” col. 9, lines 20-25) ; generating, via a trained machine learning model and based on the first state and the one or more constraints, a first action for the character to perform (“a generative control engine 120 [of Fig. 1B] which receives the combination of the character control variables 116 and user input 114. The generative control engine 120 may then output a control signal for use by the dynamic animation generation system 100 in generating animation control information 112B for frame ‘i+1’,” col. 9, lines 35-45) ; and causing the character to perform the first action within a computer-based or physical environment (“The control signal may therefore be used to output a large variation of subtle motion behaviors,” col. 9 , line 55). Regarding claim 2 , Starke discloses wherein generating the first action comprises: sampling a prior distribution based on the first state and the one or more constraints to generate a latent vector (“the system combines the user input with the character control variables to form combined input,” col. 17, lines 60-65; “the system samples the latent feature space based on the combined input,” col. 18, lines 45-50; “the generative control engine may then sample the latent feature space about the encoding, for example via the addition of noise (e.g., sampled from Gaussian distribution),” col. 9, line 65 – col. 10, line 5) ; and processing the latent vector and the first state using a controller included in the trained machine learning model to generate the first action (“The sample may then be decoded via the autoencoder and provided as the control signal,” col. 9, line 65 – col. 10, line 5). Regarding claim 4 , Starke discloses training a first machine learning model to generate the trained machine learning model (“a machine learning model may be trained based on motion capture information,” col. 2, lines 1-5) , wherein the first machine learning model comprises an encoder (“the generative control model may represent an autoencoder which encodes received user input into a learned latent feature space,” col. 7, lines 5-10). Regarding claim 5 , Starke discloses wherein the trained machine learning model comprises at least one of a trained variational autoencoder (VAE) or a trained generative model (“the generative control engine may represent an autoencoder … the generative control engine may then sample the latent feature space about the encoding, for example via the addition of noise (e.g., sampled from Gaussian distribution). The sample may then be decoded via the autoencoder and provided as the control signal,” col. 9, line 55 – col. 10, line 5). Regarding claim 6 , Starke discloses generating, via the trained machine learning model and based on the one or more constraints and a second state of the character subsequent to performing the first action (“the system combines the user input with the character control variables to form combined input,” col. 17, lines 60-65) , a second action for the character to perform; and causing the character to perform the second action within the computer-based or physical environment (“the motion prediction engine 130 [of Fig. 1] may output a character pose for frame ‘i+1’. Similar to the above, the animation control information 112B may then be provided as input to the dynamic animation generation system 100, which may continue autoregressively generating motion for the in-game character,” col. 11, lines 1-10). Regarding claim 8 , Starke discloses training a first machine learning model to produce the trained machine learning model based on a reward that is a metric of comparison between motions generated by the first machine learning model and motions sampled from a set of motion recordings (“The character control variables 206 [of Fig. 2C] may represent a ground truth, such that a reconstruction loss 238 may be determined. In some embodiments, the reconstruction loss 238 may be determined using mean squared error between the output of the decoder engine 234 and the input character control variables 206. In some embodiments, an L1 loss may be used for the reconstruction loss 238. In addition to the reconstruction loss 238, an adversarial loss 240 may optionally be determined. For example, the discriminator engine 236 may output a score associated with real and fake inputs,” col. 14, lines 40-50). Regarding claim 9 , Starke discloses wherein a controller that controls one or more joints of the character causes the character to move according to the first action (“the pose may identify positions of bones or joints of the in-game character,” col. 4, lines 40-45; “the motion generation model may cause character poses of the in-game characters to be updated such that animation may be generated,” col. 17, lines 10-15). Regarding claim 10 , Starke discloses wherein the character comprises either a virtual character or a physical robot (“the in-game character,” col. 4, lines 40-45). Regarding claim 11 , it is rejected using the same citations and rationales described in the rejection of claim 1, with the additional limitations of One or more non-transitory computer-readable media storing instructions that, when executed by at least one processor, cause the at least one processor to perform the steps (“program code can be found embodied in a tangible non-transitory signal-bearing medium,” Starke, col. 22, lines 5-10). Regarding claims 12, 14, 17, and 18 , they are rejected using the same citations and rationales described in the rejections of claims 2, 4, 8, and 9, respectively. Regarding claim 19 , Starke discloses wherein the environment is at least one of a simulation environment, an extended reality (XR) environment, a game environment, or a physical environment (“an in-game character of an electronic game,” abstract). Regarding claim 20 , it is rejected using the same citations and rationales described in the rejection of claim 1, with the additional limitations of A system, comprising: one or more memories storing instructions; and one or more processors that are coupled to the one or more memories and, when executing the instructions, are configured (Starke, Fig. 6) . Claim Rejections - 35 USC § 103 07-20-aia AIA 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. 07-21-aia AIA Claim s 3 and 13 are rejected under 35 U.S.C. 103 as being unpatentable over Starke in view of Saito et al. (US 2023/0260182; hereinafter “Saito”) . Regarding claim 3 , Starke discloses processing the first state and the one or more constraints (“the system combines the user input with the character control variables to form combined input,” col. 17, lines 60-65). Starke does not disclose using a transformer encoder to generate the prior distribution . In the same art of character animation, Saito teaches using a transformer encoder to generate the prior distribution (“the human motion generation system utilizes a neural network encoder including convolutional layers or transformer layers to generate a sequence of latent feature representations in a continuous latent space based on the digital scene,” para. 16). Before the effective filing date of the claimed invention, it would have been obvious to one having ordinary skill in the art to apply the teachings of Saito to Starke. The motivation would have been “the transformer-based architecture provides comparable or improved reconstruction” (Saito, para. 73). Regarding claim 13 , it is rejected using the same citations and rationales described in the rejection of claim 3 . 07-21-aia AIA Claim s 7, 15, and 16 are rejected under 35 U.S.C. 103 as being unpatentable over Starke in view of Shi et al. (CN 113112577; hereinafter “Shi”; machine translation used for citations) . Regarding claim 7 , Starke discloses training a first machine learning model to produce the trained machine learning model by: sampling a first motion from a set of motion recordings and a timestep within the first motion to generate a sampled motion (“access a motion capture database 202 [of Fig. 2A] and obtain motion capture information 204 for utilization. Example motion capture information 204 may include frames (e.g., image frames, time stamps of information) which may be used to identify aspects of a motion capture actor,” col. 11, lines 40-50). Starke does not disclose removing at least one joint or at least one frame within the sampled motion to generate a masked motion; generating, via the first machine learning model and based on a second state of the character and the masked motion, a second action for the character to perform; causing the character to perform the second action within the computer-based environment to reach a third state of the character; and updating one or more parameters of the first machine model based on a comparison between the third state and a fourth state of the character included in the sampled motion . In the same art of character animation, Shi teaches training a first machine learning model to produce the trained machine learning model (“a training method for a transition frame prediction model,” para. 5) by: sampling a first motion from a set of motion recordings and a timestep within the first motion to generate a sampled motion (“Acquire target joint data corresponding to multiple consecutive motion frames, wherein the multiple consecutive motion frames include keyframes and intermediate frames,” para. 7) ; removing at least one joint or at least one frame within the sampled motion to generate a masked motion (“Mask the target joint data of the intermediate frame to obtain the joint mask data of the intermediate frame,” para. 8) ; generating, via the first machine learning model and based on a second state of the character and the masked motion, a second action for the character to perform; causing the character to perform the second action within the computer-based environment to reach a third state of the character (“The joint mask data of the intermediate frame and the target joint data of the key frame are used as inputs to the multi-head self-attention module … calculating the joint prediction result of the intermediate frame using the feedforward neural network,” paras. 21-23) ; and updating one or more parameters of the first machine model based on a comparison between the third state and a fourth state of the character included in the sampled motion (“adjusting the learnable parameters of the self-attention neural network model so that the difference between the joint prediction result of the intermediate frame output by the self-attention neural network model and the target joint data of the intermediate frame is less than a threshold,” para. 28). Before the effective filing date of the claimed invention, it would have been obvious to one having ordinary skill in the art to apply the teachings of Shi to Starke. The motivation would have been to “improve the accuracy of transition frame prediction” (Shi, para. 5). Regarding claim 15 , it is rejected using the same citations and rationales described in the rejection of claim 7. Regarding claim 16 , the combination of Starke and Shi renders obvious terminating training of the first machine learning model using the sampled motion based on a similarity between the third state and the fourth state being less than a predefined threshold (“adjusting the learnable parameters of the self-attention neural network model so that the difference between the joint prediction result of the intermediate frame output by the self-attention neural network model and the target joint data of the intermediate frame is less than a threshold,” Shi, para. 28; see claim 7 for motivation to combine). Pertinent Prior Art The following prior art is considered pertinent to applicant's disclosure: Mourot et al. (“A Survey on Deep Learning for Skeleton-Based Human Animation”) discloses a review of the prior art for predicting character animation poses using machine learning and has many relevant teachings Conclusion Any inquiry concerning this communication or earlier communications from the examiner should be directed to Ryan McCulley whose telephone number is (571)270-3754. The examiner can normally be reached Monday through Friday, 8:00am - 4:30pm. Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice. If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Kee Tung can be reached at (571) 272-7794. 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. /RYAN MCCULLEY/Primary Examiner, Art Unit 2611 Application/Control Number: 18/830,218 Page 2 Art Unit: 2611 Application/Control Number: 18/830,218 Page 3 Art Unit: 2611 Application/Control Number: 18/830,218 Page 4 Art Unit: 2611 Application/Control Number: 18/830,218 Page 5 Art Unit: 2611 Application/Control Number: 18/830,218 Page 6 Art Unit: 2611 Application/Control Number: 18/830,218 Page 7 Art Unit: 2611 Application/Control Number: 18/830,218 Page 8 Art Unit: 2611 Application/Control Number: 18/830,218 Page 9 Art Unit: 2611 Application/Control Number: 18/830,218 Page 10 Art Unit: 2611 Application/Control Number: 18/830,218 Page 11 Art Unit: 2611
Prosecution Insights