DYNAMIC DETERMINATION OF HUMAN GESTURES BASED ON CONTEXT

• US 20180060706A1
• Filed: 09/01/2016
• Published: 03/01/2018
• Est. Priority Date: 09/01/2016
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: ALLEN, Corville O., CHUNG, Albert A., MUTHUKRISHNAN, Ravi K.
• Granted Patent: US 10,832,071 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 3/017 Gesture based interaction, e.g. based on a set of recognized hand gestures interaction based on gestures traced on a digitiser G06F3/04883
• CPC Class Codes: G06V 10/768 using context analysis, e.g. recognition aided by known co-occurring patterns
• CPC Class Codes: G06V 40/16 Human faces, e.g. facial parts, sketches or expressions
• CPC Class Codes: G06V 40/28 Recognition of hand or arm movements, e.g. recognition of deaf sign language static hand signs G06V40/113

Abstract

A system comprises a processor configured to execute instructions to receive an indication of an occurrence of a human gesture and to perform an analysis of the indication of the occurrence of the human gesture to determine contextual criteria having a relationship to the occurrence of the human gesture. The processor may determine a meaning of the human gesture based at least in part on the contextual criteria and a plurality of possible intended meanings for the human gesture. The processor also may execute an instruction respective to determining the meaning of the human gesture, wherein at least a portion of the instruction is dependent upon the meaning of the human gesture.

Claims

1. A computer program product for determination of a contextual meaning of human gestures, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to:

receive a visual image asset comprising a plurality of pixels that collectively depict a content of the visual image asset;

analyze the plurality of pixels according to a first digital image analysis protocol to determine at least one indicia of a context associated with the content of the visual image asset;

analyze the plurality of pixels according to a second digital image analysis protocol to determine a gesture indicated by the content of the visual image asset;

perform a semantic mapping to determine the contextual meaning of the gesture based on the at least one indicia of the context and the determined gesture; and

execute an operation responsive to determining the contextual meaning of the gesture, wherein at least one function of the operation is dependent upon the contextual meaning of the gesture.

2. The computer program product of claim 1, wherein at least one of the first digital image analysis and the second digital image analysis is selected from among a group consisting of facial recognition, clothing recognition, location identification, temporal identification, and role identification.

3. The computer program product of claim 1, wherein the at least one indicia of the context associated with the content of the visual image asset indicates a contextual characteristic selected from among a group consisting of culture, ethnicity, religion, country, region, location, setting, role, and time.

4. The computer program product of claim 1, wherein the gesture is performed by a non-human entity.

5. The computer program product of claim 1, wherein executing the program instructions further causes the processor to determine a geographic area associated with the at least one indicia of the context.

6. The computer program product of claim 5, wherein performing the semantic mapping to determine the contextual meaning of the gesture comprises:

determining one or more cultural and societal customs corresponding to the geographic area associated with the at least one indicia of the context;

determining one or more possible meanings of the gesture;

mapping the one or more cultural and societal customs to the one or more possible meanings of the gesture to determine a probability of accuracy for each of the one or more possible meanings of the gesture according to the geographic area associated with the at least one indicia of the context indicated; and

determining a possible meaning of the gesture having a highest determined probability of accuracy and selected from among the one or more possible meanings of the gestures as the contextual meaning of the gesture.

7. The computer program product of claim 1, wherein the computer program product is executed in a cloud environment as a software as a service.

8. A computer-implemented method, comprising:

receiving a digital media asset comprising a plurality of pixels;

performing a first image analysis of the digital media asset at a pixel-based level to determine a gesture embodied within the digital media asset according to a first relationship among the plurality of pixels of the digital media asset;

performing a second image analysis of the digital media asset at the pixel-based level to determine a context of the gesture according to a second relationship among the plurality of pixels of the digital media asset; and

determining a cultural meaning of the gesture based on the determined gesture and the context of the gesture.

9. The computer-implemented method of claim 8, wherein determining the cultural meaning of the gesture comprises:

determining an area associated with the context of the gesture;

determining a plurality of potential meanings of the determined gesture; and

determining the cultural meaning of the gesture based on the plurality of potential meanings of the determined gesture and the area associated with the context of the gesture.

10. The computer-implemented method of claim 9, wherein determining the cultural meaning of the gesture based on the plurality of potential meanings of the determined gesture and the area associated with the context of the gesture comprises:

determining a score for each of a plurality of context metadata values based on the content of the digital media asset;

mapping each of the plurality of context metadata values to each of the plurality of potential meanings of the determined gesture to associate an accumulated score with each of the plurality of potential meanings of the determined gesture; and

selecting the cultural meaning of the gesture as one of the plurality of potential meanings of the determined gesture from among the plurality of potential meanings of the determined gesture based at least in part on a determination that the accumulated score of the one of the plurality of potential meanings of the determined gesture exceeds a threshold.

11. The computer-implemented method of claim 8, wherein the second image analysis is selected from a group consisting of facial recognition, clothing recognition, location identification, temporal identification, and role identification.

12. The computer-implemented method of claim 11, wherein the second image analysis determines one or more facial recognition context metadata values selected from a group consisting of race, gender, and ethnicity.

13. The computer-implemented method of claim 11, wherein the second image analysis determines one or more clothing recognition context metadata values selected from a group consisting of culture, religion, and country.

14. The computer-implemented method of claim 11, wherein the second image analysis determines one or more location identification context metadata values selected from a group consisting of region, country, location, and setting.

15. The computer-implemented method of claim 8, wherein the computer-implemented method is implemented in a cloud environment as a software as a service.

16. A system, comprising a processor configured to:

receive an indication of an occurrence of a human gesture;

execute one or more instructions to perform an analysis of the indication of the occurrence of the human gesture to determine contextual criteria having a relationship to the occurrence of the human gesture;

determine a meaning of the human gesture based at least in part on the contextual criteria and a plurality of possible intended meanings for the human gesture; and

execute an instruction responsive to determining the meaning of the human gesture, wherein at least a portion of the instruction is dependent upon the meaning of the human gesture.

17. The system of claim 16, wherein the indication of the occurrence of the human gesture is a visual image asset, and wherein the contextual criteria is determined based on surroundings of the human gesture in the visual image asset.

18. The system of claim 16, wherein the plurality of possible intended meanings for the human gesture are filtered according to the contextual criteria to determine the meaning of the human gesture.

19. The system of claim 16, wherein the contextual criteria is determined according to a multistep context metadata determination system comprising facial recognition, clothing recognition, location identification, temporal identification, and role identification.

20. The system of claim 16, wherein determining the meaning of the human gesture comprises performing a semantic mapping of the human gesture to the possible intended meanings for the human gesture according to the contextual criteria.