Enhanced pose generation based on generative modeling
Systems and methods are provided for enhanced pose generation based on generative modeling. An example method includes accessing an autoencoder trained based on poses of real-world persons, each pose being defined based on location information associated with joints, with the autoencoder being trained to map an input pose to a feature encoding associated with a latent feature space. Information identifying, at least, a first pose and a second pose associated with a character configured for inclusion in an in-game world is obtained via user input, with each of the poses being defined based on location information associated with the joints and with the joints being included on a skeleton associated with the character. Feature encodings associated with the first pose and the second pose are generated based on the autoencoder. Output poses are generated based on transition information associated with the first pose and the second pose.
1. A computer-implemented method comprising:
obtaining information of at least a first pose and a second pose associated with an in-game character, each of the poses being defined based on location information associated with the plurality of joints, the joints being included on a skeleton associated with the in-game character;
generating feature encodings associated with the at least first pose and the second pose, wherein the feature encodings are associated to a latent feature space and generated via a forward pass of an autoencoder;
generating transition information from the feature encodings, wherein the transition information indicates a connection among a latent feature space between the respective feature encodings associated with the at least first pose and the second pose; and
determining one or more output poses to generate based on the transition information, wherein the one or more output poses are generated by the autoencoder and correspond to one or more transitional poses of the in-game character between the first pose to the second pose.
2. The computer-implemented method of claim 1 , further comprising:
transmitting one or more output poses generated by the autoencoder to an animation system capable of animating the in-game character.
3. The computer-implemented method of claim 1 , wherein the character is configured for inclusion in an in-game world.
4. The computer-implemented method of claim 1 , wherein the one or more output poses include a plurality of output poses, and wherein the output poses are blended.
5. The computer-implemented method of claim 1 , wherein the connection is a line in the latent feature space, and wherein the output poses correspond to samples along the line.
6. The computer-implemented method of claim 5 , wherein the samples are at a threshold distance apart.
7. The computer-implemented method of claim 1 , wherein each pose is further defined based on velocity information associated with the joints.
8. A system comprising one or more processors and non-transitory computer storage media storing instructions that when executed by the one or more processors, cause the processors to:
obtain information of at least a first pose and a second pose associated with an in-game character, each of the poses being defined based on location information associated with the plurality of joints, the joints being included on a skeleton associated with the in-game character;
generate feature encodings associated with the at least first pose and the second pose, wherein the feature encodings are associated to a latent feature space and generated via a forward pass of an autoencoder;
generate transition information from the feature encodings, wherein the transition information indicates a connection among a latent feature space between the respective feature encodings associated with the at least first pose and the second pose; and
determine one or more output poses to generate based on the transition information, wherein the one or more output poses are generated by the autoencoder and correspond to one or more transitional poses of the in-game character between the first pose to the second pose.
9. The system of claim 8 , wherein the instructions further cause the processors to:
transmit one or more output poses generated by the autoencoder to an animation system capable of animating the in-game character.
10. The system of claim 8 , wherein the character is configured for inclusion in an in-game world.
11. The system of claim 8 , wherein the one or more output poses include a plurality of output poses, and wherein the output poses are blended.
12. The system of claim 8 , wherein the connection is a line in the latent feature space, and wherein the output poses correspond to samples along the line.
13. The system of claim 12 , wherein the samples are at a threshold distance apart.
14. Non-transitory computer storage media storing instructions that when executed by a system of one or more processors, cause the processors to:
obtain information of at least a first pose and a second pose associated with an in-game character, each of the poses being defined based on location information associated with the plurality of joints, the joints being included on a skeleton associated with the in-game character;
generate feature encodings associated with the at least first pose and the second pose, wherein the feature encodings are associated to a latent feature space and generated via a forward pass of an autoencoder;
generate transition information from the feature encodings, wherein the transition information indicates a connection among a latent feature space between the respective feature encodings associated with the at least first pose and the second pose; and
determine one or more output poses to generate based on the transition information, wherein the one or more output poses are generated by the autoencoder and correspond to one or more transitional poses of the in-game character between the first pose to the second pose.
15. The computer storage media of claim 14 , wherein the instructions further cause the processor to:
transmitting one or more output poses generated by the autoencoder to an animation system capable of animating the in-game character.
16. The computer storage media of claim 14 , wherein the character is configured for inclusion in an in-game world.
17. The computer storage media of claim 14 , wherein the one or more output poses include a plurality of output poses, and wherein the output poses are blended.
18. The computer storage media of claim 14 , wherein the connection is a line in the latent feature space, and wherein the output poses correspond to samples along the line.
19. The computer storage media of claim 18 , wherein the samples are at a threshold distance apart.
20. The computer storage media of claim 14 , wherein each pose is further defined based on velocity information associated with the joints.