Coordinated gesture and locomotion for virtual pedestrians

1. A method, comprising:

generating a data model for a first conversational gesture type, by analyzing captured video data to determine motion attribute data for a plurality of conversational gestures;

upon receiving a request to splice a gesture of the first conversational gesture type into a first animation, determining a locomotion of a first virtual character, while the first virtual character is interacting with a second virtual character within the first animation;

stylizing, by operation of one or more computer processors, a gesture of the first conversational gesture type to match style criteria associated with the locomotion of the first virtual character based on the motion attribute data of the generated data model for the first conversational gesture type; and

splicing the stylized gesture into the locomotion of the first virtual character within the first animation.

2. The method of claim 1 , further comprising:

determining a gaze of the virtual character within the first animation,

wherein splicing the gesture of the first conversational gesture type into the first animation, using the generated data model, is further based on the determined gaze of the virtual character.

3. The method of claim 1 , wherein splicing the stylized gesture of the first conversational gesture type into the locomotion of the first virtual character within the first animation further comprises:

determining a locomotion of the first virtual character within the first animation;

modifying movement of the gesture to match the determined locomotion, based on generated data model.

4. The method of claim 3 , wherein splicing the stylized gesture of the first conversational gesture type into the locomotion of the first virtual character within the first animation, further comprises:

synchronizing the modified movement of the gesture to match a tempo of the determined locomotion.

5. The method of claim 4 , wherein splicing the stylized gesture of the first conversational gesture type into the locomotion of the first virtual character within the first animation, further comprises:

simulating an arm bounce of the first virtual character, based on the determined locomotion and based on the generated data model; and

splicing the simulated arm bounce into the first animation.

6. The method of claim 1 , further comprising:

identifying a conversation within animation;

determining, at each of a plurality of positions within the conversation, a gaze of the first virtual character; and

calculating, for each of the determined plurality of gazes, a respective body orientation for the first virtual character, based on the plurality of determined gazes and the generated data model.

7. The method of claim 1 , further comprising:

determining one of a plurality of splicing techniques corresponding to the first conversational gesture type,

wherein splicing the gesture of the first conversational gesture type into the locomotion of the first virtual character within the first animation is performed using the determined splicing technique.

8. The method of claim 1 , further comprising:

determining a plurality of virtual characters, inclusive of the first virtual character, that are participants to a group conversation;

determining, for each of the plurality of virtual characters, whether the respective virtual character is an addresser or an addressee in the group conversation;

determining, for each of the plurality of virtual characters, respective head and torso orientations at one or more positions within the first animation, based on the addresser and addressee determinations; and

performing one or more post-processing operations based on at least a portion of the determined head and torso orientations.

9. The method of claim 1 , further comprising:

rendering a plurality of frames of video data, based on the first animation containing the spliced gesture of the first conversational gesture type into the movement of the first virtual character.

10. The method of claim 1 , wherein first animation for at least one of (i) the gesture and (ii) the locomotion of the first virtual character, is retrieved from a motion data store.

11. The method of claim 10 , wherein the first animation was created by an animator using an animation toolkit software application.

12. The method of claim 10 , wherein the first animation comprises motion capture data created by monitoring a movement of at least one individual wearing a motion tracking suit.

13. A non-transitory computer-readable medium containing computer program code that, when executed, performs an operation, comprising:

generating a data model for a first conversational gesture type, by analyzing captured video data to determine motion attribute data for a plurality of conversational gestures;

upon receiving a request to splice a gesture of the first conversational gesture type into a first animation, determining a locomotion of a first virtual character, while the first virtual character is interacting with a second virtual character within the first animation;

stylizing a gesture of the first conversational gesture type to match style criteria associated with the locomotion of the first virtual character based on the motion attribute data of the generated data model for the first conversational gesture type; and

splicing the stylized gesture into the locomotion of the first virtual character within the first animation.

14. The non-transitory computer-readable medium of claim 13 , the operation further comprising:

determining a gaze of the virtual character within the first animation,

wherein splicing the gesture of the first conversational gesture type into the first animation, using the generated data model, is further based on the determined gaze of the virtual character.

15. The non-transitory computer-readable medium of claim 13 , wherein splicing the stylized gesture of the first conversational gesture type into the locomotion of the first virtual character within the first animation, further comprises:

determining a locomotion of the first virtual character within the first animation;

modifying movement of the gesture to match the determined locomotion, based on generated data model.

16. The non-transitory computer-readable medium of claim 15 , wherein splicing the stylized gesture of the first conversational gesture type into the locomotion of the first virtual character within the first animation, further comprises:

synchronizing the modified movement of the gesture to match a tempo of the determined locomotion.

17. The non-transitory computer-readable medium of claim 16 , wherein splicing the stylized gesture of the first conversational gesture type into the locomotion of the first virtual character within the first animation, further comprises:

simulating an arm bounce of the first virtual character, based on the determined locomotion and based on the generated data model; and

splicing the simulated arm bounce into the first animation.

18. The non-transitory computer-readable medium of claim 13 , the operation further comprising:

identifying a conversation within animation;

determining, at each of a plurality of positions within the conversation, a gaze of the first virtual character; and

calculating, for each of the determined plurality of gazes, a respective body orientation for the first virtual character, based on the plurality of determined gazes and the generated data model.

19. The non-transitory computer-readable medium of claim 10 , the operation further comprising:

determining a plurality of virtual characters, inclusive of the first virtual character, that are participants to a group conversation;

determining, for each of the plurality of virtual characters, whether the respective virtual character is an addresser or an addressee in the group conversation;

determining, for each of the plurality of virtual characters, respective head and torso orientations at one or more positions within the first animation, based on the addresser and addressee determinations and an updated position of the first virtual character, the updated position determined using the generated data model and based on the locomotion of the first virtual character within the animation; and

performing one or more post-processing operations based on at least a portion of the determined head and torso orientations.

20. A system, comprising:

one or more computer processors; and

a memory containing computer program code that, when executed by operation of the one or more computer processors, performs an operation comprising:

generating a data model for a first conversational gesture type, by analyzing captured video data to determine motion attribute data for a plurality of conversational gestures;

upon receiving a request to splice a gesture of the first conversational gesture type into a first animation, determining a locomotion of a first virtual character, while the first virtual character is interacting with a second virtual character within the first animation;

stylizing a gesture of the first conversational gesture type to match style criteria associated with the locomotion of the first virtual character based on the motion attribute data of the generated data model for the first conversational gesture type; and

splicing the stylized gesture into the locomotion of the first virtual character within the first animation.