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 .
DETAILED ACTION
Status of Claims
Claims 1–19 are pending in the application. Claim 6 is allowed. Claims 1-4, 7-9, 11-13, 16-18 are rejected. Claim 20 is canceled.
Claims 5, 10, 14, 15, and 19 are objected to.
Allowable Subject Matter
Claims 5, 10, 14, 15, and 19 are objected to as being dependent upon a rejected base claim(s), but would be allowable if rewritten in independent form including all of the limitations of the base claim(s) and any intervening claim(s).
Overview of Grounds of Rejection
Ground of Rejection
Claim(s)
Statute(s)
Reference(s)
1
1, 11
§ 103
Aziz et al. (US20210248804A1); Corazza et al. (US20120019517A1); Ahuja et al. (NPL); Wilson et al. (US20180091732A1)
2
3, 4, 12, 13
§ 103
Aziz et al. (US20210248804A1); Corazza et al. (US20120019517A1); Ahuja et al. (NPL); Wilson et al. (US20180091732A1); Li et al. (NPL)
3
2, 9, 18
§ 103
Aziz et al. (US20210248804A1); Corazza et al. (US20120019517A1); Ahuja et al. (NPL); Wilson et al. (US20180091732A1); Monroy-Hernández et al. (US20210097745A1)
4
7, 16
§ 103
Aziz et al. (US20210248804A1); Corazza et al. (US20120019517A1); Ahuja et al. (NPL); Wilson et al. (US20180091732A1); Hong et al. (NPL)
5
8, 17
§ 103
Aziz et al. (US20210248804A1); Corazza et al. (US20120019517A1); Ahuja et al. (NPL); Wilson et al. (US20180091732A1); Borke et al. (US20160267699A1)
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.
(Please refer to the paragraph(s) mentioned or nearby paragraphs of the references for the paraphrased text.)
Claims 1, 11 are rejected under 35 U.S.C. § 103 as being unpatentable over Aziz (US20210248804A1) in view of Corazza et al (US20120019517A1), further in view of Ahuja (NPL), and still further in view of Wilson et al. (US20180091732A1).
As per Claim 1, Aziz teaches the following portion of Claim 1, which recites:
“A computer-implemented method, performed by at least one processor, for personalized avatar generation, the method comprising:
receiving a text prompt describing an avatar pose;”
Aziz describes a method in which an electronic device:
“receiv[es] text, determin[es] an emotional state, and generat[es], using a neural network, a speech data set representing the received text and a set of parameters representing one or more movements of an avatar based on the received text and the determined emotional state.” – Aziz et al., ¶ [0005]
This passage shows the system receives text that is then used to drive “one or more movements of an avatar”. A person of ordinary skill in the art (POSITA) would recognize that such received text is a text prompt describing how the avatar should move or be posed.
Aziz alone does not explicitly teach all the limitation(s) of the claim. However, when combined with Ahuja et al. (NPL), they collectively teach some of the limitation(s).
Aziz and Ahuja explicitly teach the following portion of Claim 1, which recites:
“generating a first pose based on the text prompt using a first model trained to receive representations of natural language pose descriptions and, in response, produce avatar pose data;”
Aziz explains that the neural network produces motion parameters from text and is trained on text–motion pairs:
“…generating, using a neural network, a speech data set representing the received text and a set of parameters representing one or more movements of an avatar based on the received text and the determined emotional state.” – Aziz et al., ¶ [0005]
“Prior to receiving text 802 and determining speech data set 812 and animation parameters 814, neural network 810 is trained… To train neural network 810, a set of training text data is provided… along with… a reference set of parameters representing one or more movements of an avatar.” – Aziz et al., ¶ [0281]
These passages show a first model (neural network 810) that, when given text, generates “a set of parameters representing one or more movements of an avatar” – that is, avatar pose data.
Ahuja et al (Language2Pose) further confirms that training models on natural language pose descriptions to output pose sequences was well known. Ahuja describes:
“…we address this multimodal problem by introducing a neural architecture called Joint Language-to-Pose (or JL2P), which learns a joint embedding of language and pose.” – Ahuja et al (NPL), Abstract, p.1
“As an example, consider a natural language sentence which describes a human’s motion: ‘A person walks in a circle’. The goal of this cross-modal language-to-pose translation task is to generate an animation representing the sentence; i.e. an animation that shows a person following a trajectory of a circle with a walking motion.” – Ahuja et al (NPL), Sec. 3, “Problem Statement”, p.3
“Given an input sentence, an animation can be sampled from this model at inference stage.” – Ahuja et al (NPL), Sec. 4, “Joint Language-to-Pose”, p.4
Ahuja therefore shows a model that is trained on natural language motion/pose descriptions and outputs pose sequences (animations). Aziz alone shows text-to-motion parameters; when combined with Ahuja, a POSITA would understand that the claimed first model is a text-to-pose model trained to receive representations of natural language pose descriptions and produce avatar pose data, as in the claim.
Aziz and Ahuja alone do not explicitly teach all the limitation(s) of the claim. However, when combined with Corazza, they collectively teach some of the limitation(s).
Corazza explicitly teach the following portion of Claim 1, which recites:
“re-targeting the first pose, using pose data from the first model, to a target avatar body;”
Corazza discloses an autorigging and animation method that includes retargeting motion data:
“In another embodiment, a method of animating a 3D character includes: automatically rigging at least one mesh defining the external appearance of a 3D character… characterizing the fitted skeleton with respect to a reference skeleton, retargeting motion data defined with respect to the reference skeleton to the fitted skeleton of the 3D character, and animating the 3D character by driving the at least one mesh in accordance with the fitted skeleton, the skinning weights, and the retargeted motion data.” – Corazza et al, ¶ [0018]
Here, “motion data defined with respect to the reference skeleton” corresponds to pose data (e.g., the first pose parameters generated by the Aziz/Ahuja model), and “retargeting… to the fitted skeleton of the 3D character” is a direct match to re-targeting the first pose to a target avatar body.
Aziz and Ahuja provide the first pose data; Corazza shows how such motion data is retargeted to a particular character’s skeleton.
Aziz, Ahuja, and Corazza alone do not explicitly teach all the limitation(s) of the claim. However, when combined with Wilson, they collectively teach all of the limitation(s).
Wilson explicitly teaches the following portion of Claim 1, which recites:
“identifying a predefined avatar configuration corresponding to a user based on a profile of the user;”
Wilson’s system shows user-specific placeholder avatars and their feature configurations:
“…an electronic device with a display and one or more input devices displays an avatar editing user interface that includes: an avatar having a plurality of editable features; a feature-selection control region that includes representations of a plurality of avatar features, including a first avatar feature that is currently selected for modification; and a feature-option control region that includes representations of a plurality of options for the first avatar feature.” – Wilson et al, ¶ [0008]
Wilson further explains:
“In some embodiments the electronic device generates the user-specific placeholder avatar (e.g., 624) based the first image data and using a predetermined skin tone color independent of the first image data (e.g., generating a black and white avatar that has shapes and features that resemble the user in the first image data without any skin tone coloring); and displays the user-specific placeholder avatar prior to receiving the user input confirming the respective suggested skin tone color.” – Wilson et al, ¶ [0231]
These passages show that Wilson derives a user-specific placeholder avatar (with default attributes) from user-related data (image data) and suggested attributes. That corresponds to identifying a predefined avatar configuration that corresponds to a user and is based on their “profile” (captured image and associated attribute suggestions).
Wilson et al. explicitly teach the following portion of Claim 1, which recites:
“converting the target avatar body by applying the predefined avatar configuration to the target avatar body;”
Wilson describes applying selected feature options to the avatar:
“…updating the feature-option control region to indicate that the second option for the first avatar feature is currently selected; and updating the avatar to change the appearance of the first avatar feature in accordance with the second option for the first avatar feature.” – Wilson et al, ¶ [0008]–[0009]
By “updating the avatar to change the appearance” according to selected options, Wilson is effectively applying a configuration of features to the avatar body. That is, the target avatar body is “converted” by applying the chosen avatar configuration (feature options).
Corazza and Wilson explicitly teach the following portion of Claim 1, which recites:
“rendering an avatar, using a second model, based on the target avatar body with the predefined avatar configuration, wherein the avatar is in the first pose.”
Corazza’s method, quoted above, shows a distinct animation stage that uses retargeted motion data to drive the character mesh:
“…retargeting motion data defined with respect to the reference skeleton to the fitted skeleton of the 3D character; and animating the 3D character by driving the at least one mesh in accordance with the fitted skeleton, the skinning weights, and the retargeted motion data.” – Corazza et al, ¶ [0018]
This animation stage functions as a second model or module that consumes pose data and produces a visual animation of the avatar in that pose.
Wilson provides the rendering of an avatar that has already been configured. For example, when discussing avatar 901:
(paraphrased from figure description) avatar 901 is displayed on a display of the electronic device, updated to reflect new feature selections such as hair style and color – Wilson et al, ¶¶ [0266]–[0268] (avatar 901 in FIGS. 9B–9D)
Together, Corazza and Wilson show that a configured avatar body (with user-specific configuration applied) can be driven by a motion/pose engine and rendered on screen. That corresponds to rendering an avatar, using a second model, based on the target avatar body with the predefined avatar configuration, wherein the avatar is in the first pose.
Before the effective filing date of the claimed invention, a person of ordinary skill in the art would have been motivated to combine Aziz et al, Ahuja et al (NPL), Corazza et al, and Wilson et al to achieve the claimed method. Aziz shows that natural-language text can be used to drive avatar movements via a neural network that outputs motion parameters from text. Ahuja demonstrates that models can be trained specifically on natural language pose descriptions to produce sequences of 3D human poses, confirming that a language-to-pose training regime is standard practice. Corazza provides a robust rigging and retargeting pipeline that applies motion data to arbitrary character skeletons and animates those characters, thereby enabling reuse of pose data across different avatar bodies. Wilson contributes a user-specific avatar configuration framework that defines and applies personalized configurations to avatars and renders them on a display. Combining these teachings would predictably yield a system that (1) uses natural-language pose descriptions to generate pose data, (2) retargets that pose data to a chosen avatar body, (3) applies a user-specific configuration to that body, and (4) renders the personalized avatar in the requested pose. This combination improves flexibility and realism in personalized avatar generation, and would be seen by a POSITA as a straightforward integration of known components with no unexpected results.
PNG
media_image1.png
13
460
media_image1.png
Greyscale
System Claim 11 does not include any additional limitations that would significantly distinguish them from method claim 1. Therefore, it is likewise rejected under 35 U.S.C. § 103 in view of the same references and for the same reasons set forth above. Aziz teaches using a memory storing program instructions and processor for executing the program instruction in [0007].
PNG
media_image1.png
13
460
media_image1.png
Greyscale
Claims 3, 4, 12, 13 are rejected under 35 U.S.C. § 103 as being unpatentable over Aziz (US20210248804A1) in view of Corazza et al (US20120019517A1), further in view of Ahuja (NPL), still further in view of Wilson et al. (US20180091732A1), and still further in view of Li et al. (NPL)
As per Claim 3, Aziz alone does not explicitly teach all the limitation(s) of the claim. However, when combined with Li, they collectively teach all the limitation(s).
Li teaches the limitation(s) of Claim 3 that recites:
“The computer-implemented method of claim 1, wherein the first pose is a 3D pose generated based on human joint and limb orientation and positioning parameters.”
Li explicitly discloses generation of detailed 3D human poses using joint and limb orientation parameters:
"We represent dance as a 3D motion sequence that consists of joint rotation and global translation, which enables easy transfer of our output for applications such as motion retargeting." — Li et al. Section(s): Introduction
"We recover the camera calibration parameters and the 3D human motion in terms of SMPL parameters." — Li et al. Section(s): 3. AIST++ Dataset
Thus, Li meets the requirements of Claim 3 by clearly disclosing a "3D pose generated based on human joint and limb orientation and positioning parameters."
It would have been obvious for a person of ordinary skill in the art, before the effective filing date of the claimed invention, to combine the detailed joint and limb orientation-based 3D pose generation explicitly taught by Li with the text-driven avatar animation methods explicitly disclosed by Aziz and Wilson. Aziz and Li explicitly disclose the use of textual prompts for avatar pose generation, while Wilson explicitly discloses applying predefined user-specific parameters. Incorporating Li’s explicit teaching of precise 3D joint and limb orientations into the combined Aziz/Li/Wilson system represents a predictable, advantageous, and straightforward integration. This combination allows for richer, more realistic, and precisely controlled 3D avatar pose generation and animation, meeting clear industry demands for realistic avatar animation. The resulting combination yields predictable outcomes without unexpected results.
PNG
media_image1.png
13
460
media_image1.png
Greyscale
As per Claim 4, Aziz alone does not explicitly teach all the limitation(s) of the claim. However, when combined with Li, they collectively teach all the limitation(s).
Li teaches the limitation(s) of Claim 4 that recites:
“The computer-implemented method of claim 1, wherein the first pose is a human body pose represented by (Skinned Multi-Person Linear Model) SMPL parameters.”
Li explicitly states:
"We recover the camera calibration parameters and the 3D human motion in terms of SMPL parameters." — Li et al. Section(s): 3. AIST++ Dataset
This disclosure provided by Li clearly and directly matches the requirement of Claim 4, which explicitly requires a "human body pose represented by SMPL parameters."
Before the effective filing date of claimed invention, it would have been obvious to a person of ordinary skill in the art at the time of invention to integrate Li’s explicit disclosure of human body poses represented by SMPL parameters into the avatar generation framework explicitly taught by Aziz and Wilson. Aziz and Li explicitly disclose generating and retargeting avatar poses from textual prompts, and Wilson explicitly teaches applying predefined user-specific avatar parameters. Combining these references with Li’s explicit use of SMPL parameters for precise human body representation would be straightforward and predictable, providing improved realism and controllability of avatar poses and animations. Such combination represents a logical design choice for enhancing the accuracy and realism of avatar generation, and would yield no unexpected results.
PNG
media_image1.png
13
460
media_image1.png
Greyscale
System Claim 12 does not include any additional limitations that would significantly distinguish them from method claim 3. Therefore, it is likewise rejected under 35 U.S.C. § 103 in view of the same references and for the same reasons set forth above.
PNG
media_image1.png
13
460
media_image1.png
Greyscale
System Claim 13 does not include any additional limitations that would significantly distinguish them from method claim 4. Therefore, it is likewise rejected under 35 U.S.C. § 103 in view of the same references and for the same reasons set forth above.
PNG
media_image1.png
13
460
media_image1.png
Greyscale
Claims 2, 9, 18 are rejected under 35 U.S.C. § 103 as being unpatentable over Aziz (US20210248804A1) in view of Corazza et al (US20120019517A1), further in view of Ahuja (NPL), still further in view of Wilson et al. (US20180091732A1), and still further in view of Monroy-Hernández et al. (US20210097745A1).
As per Claim 2, Aziz alone does not explicitly teach all of the limitation(s) of the claim. However, when combined with Monroy-Hernández et al., they collectively teach all the limitation(s).
Monroy-Hernández teaches the limitation(s) of Claim 2 that recites:
“The computer-implemented method of claim 1, wherein the text prompt further describes a scene and interactions of the avatar within the scene.”
Monroy-Hernández explicitly discloses avatar scenarios involving both a scene and interactions of avatars:
“The avatar story typically progresses through one or more scenes involving action or dialog by or between user avatars.” — Monroy-Hernández et al., [0014]
This disclosure aligns with Claim 2, clearly establishing that Monroy-Hernández teaches avatar scenes ("one or more scenes") and interactions ("action or dialog by or between user avatars").
While Monroy-Hernández explicitly uses story templates rather than free-form textual prompts, Aziz and Li explicitly teach the use of direct textual prompts for driving avatar generation.
Combining Monroy-Hernández’s explicit teaching of scene interactions with the explicit textual prompt methods of Aziz and Li would have been an obvious design choice to POSITA before the effective filing of the claimed invention, enabling Aziz’s and Li’s systems to explicitly describe scenes and interactions within free-form textual prompts. Such integration involves substituting Aziz’s and Li’s existing input methods for Monroy-Hernández’s templates to achieve the predictable outcome of enhanced avatar interactivity within user-described scenes, meeting clear industry demands for richer and more personalized avatar experiences without yielding unexpected results.
PNG
media_image1.png
13
460
media_image1.png
Greyscale
As per Claim 9, Aziz does not explicitly teach all of the limitation(s) of the claim. However, when combined with Monroy-Hernández et al., they collectively teach all the limitation(s).
Monroy-Hernández teaches the limitation(s) of Claim 9 that recites:
"The computer-implemented method of claim 1, further comprising generating an image, using the second model, of a scene in a virtual environment including the avatar interacting with objects in the scene."
Monroy-Hernández explicitly discloses generating scene images explicitly including avatars interacting explicitly with scene objects:
Figure 5 explicitly teaches at step 510 the generation of virtual scene images explicitly based on user input explicitly provided in step 502 (second model input).
"Using the user inputs 502 and the real-time data 504, the parameterized avatar system 130 may then assemble the specific media assets required by the story template (e.g. images of skylines or landmarks, customized story characters, relevant objects etc.), populate the story template, and transmit a substantially complete parameterized avatar story for presentation to the user on the client device 102." — Monroy-Hernández et al., [0111]
Figure 8 explicitly visually depicts a generated avatar explicitly interacting with objects explicitly within a virtual scene environment.
Monroy-Hernández’s explicit disclosures directly meet Claim 9’s explicit limitations of generating an image of a virtual environment explicitly including avatar-object interactions.
Before the effective filing date of the claimed invention, it would have been obvious to a person of ordinary skill in the art (POSITA) to incorporate Monroy-Hernández’s explicit teaching of generating virtual scene images explicitly depicting avatars explicitly interacting with objects into the avatar generation and rendering systems explicitly taught by Aziz, Li, and Wilson. Aziz and Li explicitly disclose avatar pose generation and retargeting from textual prompts, while Wilson explicitly teaches avatar personalization through selectable user-specific parameters. Integrating Monroy-Hernández’s explicitly disclosed techniques for scene and object interaction into Aziz's, Li's, and Wilson’s methods would have provided predictable and straightforward enhancements to realism and immersive user experiences. Such combination directly addresses known demands for richer, personalized avatar interactions without yielding unexpected results.
PNG
media_image1.png
13
460
media_image1.png
Greyscale
System Claim 18 does not include any additional limitations that would significantly distinguish them from method claim 9. Therefore, it is likewise rejected under 35 U.S.C. § 103 in view of the same references and for the same reasons set forth above.
PNG
media_image1.png
13
460
media_image1.png
Greyscale
Claims 7, 16 are rejected under 35 U.S.C. § 103 as being unpatentable over Aziz (US20210248804A1) in view of Corazza et al (US20120019517A1), further in view of Ahuja (NPL), still further in view of Wilson et al. (US20180091732A1), and still further in view of Hong et al. (NPL).
As per Claim 7, Aziz alone does not explicitly teach all of the limitation(s) of the claim. However, when combined with Hong et al. (NPL), they collectively teach all the limitation(s).
Hong et al. teach the limitation(s) of Claim 7 that recites:
"The computer-implemented method of claim 1, wherein the target avatar body is a gray avatar-human body representation."
Hong discloses the claimed limitation visually in Figure 1, which clearly illustrates a neutral, explicitly gray-colored intermediate human-avatar mesh representation. Specifically, Figure 1 shows:
A baseline SMPL-based avatar depicted explicitly as a neutral, gray-colored human mesh prior to texture mapping or applying colors.
This explicit visual depiction exactly matches the limitation of Claim 7.
Before the effective filing date of the claimed invention, it would have been obvious to a person of ordinary skill in the art, at the time of invention, to utilize Hong’s visual teaching (Figure 1) of a neutral, gray-colored intermediate avatar representation within the textual-prompt-driven avatar generation systems explicitly taught by Aziz, Li, and Wilson. Such gray intermediate representations are widely known baseline models used to clearly illustrate and facilitate retargeting, customization, and texture mapping steps. Combining Hong’s explicit depiction of a neutral intermediate avatar body with Aziz's, Li's, and Wilson’s established methods for avatar pose retargeting and personalization would have been predictable and straightforward, yielding expected improvements in flexibility, efficiency, and accuracy during avatar customization and rendering processes without unexpected results.
PNG
media_image1.png
13
460
media_image1.png
Greyscale
System Claim 16 does not include any additional limitations that would significantly distinguish them from method claim 7. Therefore, it is likewise rejected under 35 U.S.C. § 103 in view of the same references and for the same reasons set forth above.
PNG
media_image1.png
13
460
media_image1.png
Greyscale
Claims 8, 17 are rejected under 35 U.S.C. § 103 as being unpatentable over Aziz (US20210248804A1) in view of Corazza et al (US20120019517A1), further in view of Ahuja (NPL), still further in view of Wilson et al. (US20180091732A1), and still further in view of Borke et al (US20160267699A1).
As per Claim 8, Aziz does not explicitly teach all of the limitation(s) of the claim. However, when combined with Corazza et al (US20120019517A1) and Borke et al (US20160267699A1), they collectively teach all the limitation(s).
Corazza and Borke teach the limitation(s) of Claim 8 that recites:
“The computer-implemented method of claim 1, wherein the re-targeting comprises matching corresponding joints, in position and orientation, from the first pose and the target avatar body.”
Corazza explicitly discloses joint-position mapping for retargeting:
"Retargeting can be used to adapt the motion data for one skeleton to animate another skeleton to accommodate variations in the number of bones and/or skeleton hierarchy and/or bone length. Retargeting of motion data generally involves characterization, which is the process of bringing the character to a reference pose and of mapping the joints of the character's skeleton to a predefined reference skeleton." — Corazza et al., [0049]
Borke explicitly discloses matching joint orientations explicitly for retargeting:
"The avatar mapping process first calculates each joint’s direction by finding the rotation from its initial direction (the unit vector from this joint’s starting position to the initial position of this joint’s first child) to its current direction… Then, for joints that can rotate about the axis of its own direction (roll), the mapping process calculates the change in angle between the initial roll vector and current roll vector… joints are given their new rotations…" — Borke et al., [0118] –[0119]
These disclosures clearly and fully satisfy Claim 8’s limitation of matching corresponding joints explicitly in both "position" (Corazza) and "orientation" (Borke).
Before the effective filing date of the claimed invention, it would have been obvious to a person of ordinary skill in the art, at the time of invention, to combine Corazza’s explicit joint-position mapping technique with Borke’s explicit joint-orientation calculation method into the avatar generation and retargeting system explicitly taught by Aziz, Li, and Wilson. Aziz and Li explicitly disclose generating and retargeting avatar poses based on textual prompts, while Wilson explicitly teaches personalized avatar appearance. Integrating Corazza’s positional matching method with Borke’s orientation matching method would be a straightforward, predictable, and logical step for a skilled artisan seeking precise and accurate retargeting results. This combination addresses clear industry demands for realistic and precisely controlled avatar animations without yielding any unexpected outcomes.
PNG
media_image1.png
13
460
media_image1.png
Greyscale
System Claim 17 does not include any additional limitations that would significantly distinguish them from method claim 8. Therefore, it is likewise rejected under 35 U.S.C. § 103 in view of the same references and for the same reasons set forth above.
PNG
media_image1.png
13
460
media_image1.png
Greyscale
Response to Arguments
Applicant’s arguments, filed on 09/22/2025, have been fully considered and are found to be persuasive. Therefore, the rejection has been withdrawn. However, upon further consideration, a new ground(s) of rejection is made on this final action.
Conclusion
The prior art made of record and relied upon in this action is as follows:
Patent Literature:
Aziz et al. (US20210248804A1), "Using Text for Avatar Animation"
Wilson et al. (US20180091732A1), "Avatar Creation and Editing"
Monroy-Hernández et al. (US20210097745A1), "Dynamic Parameterized User Avatar Stories"
Corazza et al. (US20120019517A1), "Automatic Generation of 3-D Character Animation"
Borke et al. (US20160267699A1), "Avatar Control System"
Non-Patent Literature (NPL):
Li et al. (NPL), (2021), "AI Choreographer: Music Conditioned 3D Dance Generation with AIST++"
Hong et al. (NPL), (July 2022), "AvatarCLIP: Zero-Shot Text-Driven Generation and Animation of 3D Avatars"
Ahuja et al. (NPL), (27 Nov 2019), "Language2Pose: Natural Language Grounded Pose Forecasting"
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure and is listed as follows:
Patent Literature:
Black et al. (US20190392626A1), "Skinned Multi-Person Linear Model (SMPL)"
Hwang et al. (US20120139830A1), "Apparatus and Method for Controlling Avatar Using Expression Control Point"
Non-Patent Literature (NPL):
Kim et al. (NPL), (2022), "Music-Conditioned Pluralistic Dancing Composition with a Transformer-based GAN"
Zhang et al. (NPL), (2022), "Dance Generation with Style Embedding: Learning the Latent Representations of Dance Styles"
Ferreira et al. (NPL), (2021), "Learning to Dance: A Graph Convolutional Adversarial Network to Generate Realistic Dance Motions from Audio"
Aberman et al. (NPL), (2020), "Unpaired Motion Style Transfer from Video to Animation"
Delmas et al. (NPL), (21 Oct 2022), "PoseScript: 3D Human Poses from Natural Language"
Zhang et al. (NPL), (2020), "Adversarial Synthesis of Human Pose from Text"
THIS ACTION IS MADE FINAL. 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 ADEEL BASHIR whose telephone number is (571) 272-0440. The examiner can normally be reached Monday-Thursday.
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, Daniel Hajnik can be reached on (571) 276-7642. 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.
/ADEEL BASHIR/
Examiner, Art Unit 2616
/DANIEL F HAJNIK/Supervisory Patent Examiner, Art Unit 2616