Generating and using a predictive virtual personification

• US 9,727,798 B2
• Filed: 07/20/2015
• Issued: 08/08/2017
• Est. Priority Date: 07/18/2014
• Status: Active Grant

• Current Assignee: Acrovirt, LLC
• Original Assignee: Acrovirt, LLC
• Inventors: Caldwell, James Michael
• Primary Examiner(s): Williams, Kimberly A
• Application Number:
  US14/804,302
• Publication Number:
  US 20160019434A1
• Time in Patent Office: 750 Days
• Field of Search: 345474, 382159, 382170, 382171, 382190
• CPC Class Codes: A63F 2300/6607 for animating game characters, e.g. skeleton kinematics
• CPC Class Codes: G06F 18/214 Generating training patterns; Bootstrap methods, e.g. bagging or boosting
• CPC Class Codes: G06V 10/462 Salient features, e.g. scale invariant feature transforms [SIFT]
• CPC Class Codes: G06V 40/23 Recognition of whole body movements, e.g. for sport training

Abstract

A system for generating a predictive virtual personification includes receiving, from an AV data source, a data store, and a saliency recognition engine, wherein the AV data source is configured to transmit one or more AV data sets to the saliency recognition engine, each AV data set includes a graphical representation of a donor subject, and the saliency recognition engine is configured to receive the AV data set and one or more identified trigger stimulus events, locate a set of saliency regions of interest (SROI) within the graphical representation of the donor subject, generate a set of SROI specific saliency maps and store, in the data store, a set of correlated SROI specific saliency maps generated by correlating each SROI specific saliency map a corresponding trigger event.

Claims

1. A method for generating a predictive virtual personification comprises:

receiving, from an AV data source, one or more AV data sets;

locating within each AV data set, with a saliency recognition engine, a graphical representation of a donor subject and a set of saliency regions of interest (SROI) within said graphical representation of the donor subject;

identifying one or more trigger stimulus events, wherein each trigger stimulus events precedes or is contemporaneous with one or more SROI specific reactive responses and each SROI specific reactive response is observable within a SROI;

identifying a set of donor-specific physio-emotional characteristics corresponding to a donor-specific physio-emotional state at the time of the trigger stimulus event and tagging the set of correlated SROI specific saliency maps with the corresponding set of donor-specific physio-emotional characteristics;

generating, for each SROI, a set of SROI specific saliency maps, wherein each SROI specific saliency map plots a change in geospatial orientation of one or more SROIs within a predetermined time-frame corresponding to each trigger stimulus event; and

storing, in a data store, a set of correlated SROI specific saliency maps generated by correlating each SROI specific saliency map a corresponding trigger event.

2. The method of claim 1, wherein the identifying a set of donor-specific physio-emotional characteristics comprises manually entering, with a user input device, one or more physio-emotional characteristic tags corresponding to a known physio-emotional state of the donor subject.

3. The method of claim 1, wherein the identifying a set of donor-specific physio-emotional characteristics comprises matching, with a predictive virtual presentation (PVP) correlation engine, one or more correlated SROI specific saliency maps with a plurality of historical SROI specific saliency maps, wherein each historical SROI specific saliency map corresponds to a set of physio-emotional characteristics.

4. The method of claim 3, wherein the matching the one or more correlated SROI specific saliency maps with a plurality of historical SROI specific saliency maps comprises applying, with the PVP correlation engine, a renormalization group transformation to each historical specific saliency map to generate a predictive saliency map space.

5. The method of claim 1, wherein the set of donor-specific physio-emotional characteristics comprises mood, level of rest, level of stress, or health.

6. The method of claim 1, further comprising generating, with a graphical rendering engine, an animated representation of the donor subject using the AV data set.

7. The method of claim 6, further comprising exposing the animated representation of the donor subject to a secondary stimulus event and rendering for each SROI, with a PVP rendering engine, a predicted reactive response.

8. The method of claim 7, wherein the rendering a predicted reactive response comprises identifying a secondary set of physio-emotional characteristics corresponding to the animated representation of the donor subject;

identifying one or more trigger stimulus events corresponding to the secondary stimulus event;

receiving, from the data store, each set of correlated SROI specific saliency maps corresponding to each identified trigger stimulus event and to the identified set of physio-emotional characteristics; and

generating, with the PVP rendering engine, a set of predictive SROI-specific saliency maps based on a probabilistic extrapolation as a function of the correlated SROI specific saliency maps, the identified physio-emotional characteristics, and the identified trigger stimulus event.

9. The method of claim 8, wherein the generating a set of predictive SROI-specific saliency maps comprises collecting the correlated SROI specific saliency maps into a historical saliency map space and applying a renormalization group transformation to the historical saliency map space to generate a predictive saliency map space.

10. The method of claim 8, further comprising rendering, with the graphical rendering engine, a geospatial movement of the animated representation of the donor subject by applying the set of predictive SROI-specific saliency maps to each SROI within the animated representation of the donor subject.

11. A system for generating a predictive virtual personification comprises:

an AV data source, a data store, and a saliency recognition engine;

wherein the AV data source is configured to transmit one or more AV data sets to the saliency recognition engine, each AV data set comprising a graphical representation of a donor subject; and

the saliency recognition engine comprises a non-volatile computer readable media with a computer program stored thereon, the computer program configured to receive the AV data set and one or more identified trigger stimulus events, wherein each identified trigger stimulus events precedes or is contemporaneous with one or more saliency regions of interest (SROI) specific reactive responses and each SROI specific reactive response is observable within an SROI;

locate a set of SROI within the graphical representation of the donor subject;

generate, for each SROI, a set of SROI specific saliency maps, wherein each SROI specific saliency map plots a change in geospatial orientation of one or more SROIs within a predetermined time-frame corresponding to each trigger stimulus event;

identify a set of donor-specific physio-emotional characteristics corresponding to a donor-specific physio-emotional state at the time of the trigger stimulus event and tagging the set of correlated SROI specific saliency maps with the corresponding set of donor-specific physio-emotional characteristics; and

store, in the data store, a set of correlated SROI specific saliency maps generated by correlating each SROI specific saliency map a corresponding trigger event.

12. The system of claim 11, wherein the AV data source is a historical archive comprising subsequently captured film, video, or audio data.

13. The system of claim 11, wherein the AV data source is a video camera, a television camera, a high frame rate video camera, a high resolution video camera, a motion capture device (MOCAP), or a functional imaging modality.

14. The system of claim 11, wherein the saliency recognition engine is further configured to identify a set of donor-specific physio-emotional characteristics corresponding to a donor-specific physio-emotional state at the time of the trigger stimulus event and tag the set of correlated SROI specific saliency maps with the corresponding set of donor-specific physio-emotional characteristics.

15. The system of claim 14, further comprising a graphical rendering engine configured to generate an animated representation of the donor subject based on the AV data set.

16. The system of claim 15, further comprising a predictive virtual presentation (PVP) rendering engine configured to generate a predicted reactive response to a secondary stimulus event.

17. The system of claim 16, wherein the PVP rendering engine is further configured to:

identify a secondary set of physio-emotional characteristics corresponding to the animated representation of the donor subject;

identify one or more trigger stimulus events corresponding to the secondary stimulus event;

receive, from the data store, each set of correlated SROI specific saliency maps that correspond to each identified trigger stimulus event and to the identified set of physio-emotional characteristics; and

generate a set of predictive SROI-specific saliency maps based on a probabilistic extrapolation as a function of the correlated SROI specific saliency maps, the identified physio-emotional characteristics, and the identified trigger stimulus event.

18. The system of claim 17, wherein the graphical rendering engine is further configured to render geospatial movement of the animated representation of the donor subject by applying the set of predictive SROI-specific saliency maps to each SROI within the animated representation of the donor subject.

19. The system of claim 18, further comprising an AV output device configured to project the animated representation of the donor subject into a geospatial environment.