Methods and systems for measuring group behavior

Patent metadata

Field	Value
Publication number	US-9443144-B2
Application number	US-201314080446-A
Country	US
Kind code	B2
Filing date	Nov 14, 2013
Priority date	Mar 15, 2013
Publication date	Sep 13, 2016
Grant date	Sep 13, 2016

Abstract

Methods and systems for measuring group behavior are provided. Group behavior of different groups may be measured objectively and automatically in different environments including a dark environment. A uniform visible signal comprising images of members of a group may be obtained. Facial motion and body motions of each member may be detected and analyzed from the signal. Group behavior may be measured by aggregating facial motions and body motions of all members of the group. A facial motion such as a smile may be detected by using the Fourier Lucas-Kanade (FLK) algorithm to register and track faces of each member of a group. A flow-profile for each member of the group is generated. Group behavior may be further analyzed to determine a correlation of the group behavior and the content of the stimulus. A prediction of the general public's response to the stimulus based on the analysis of the group behavior is also provided.

First claim

What is claimed is:

1. A method of measuring group behavior, comprising:
obtaining a signal comprising a sequence of images of a group having a plurality of members, wherein the group is exposed to a stimulus;
detecting a response of each member of the group to the stimulus over a period;
generating an aggregate response to the stimulus by aggregating the response of each member of the group; and
generating a group behavior measurement by determining an amount of synchrony of the response of each member relative to the aggregate response, wherein the step of generating the group behavior comprises:
generating a similarity matrix by determining a pair-wise correlation between two members of the group for a segment of the period; and
determining an entropy of the similarity matrix to measure the group behavior.

2. The method of claim 1 , further comprising predicting a second group's response to the stimulus based on the group behavior.

3. The method of claim 2 , wherein the second group's response is a crowd-sourced rating of the stimulus.

4. The method of claim 3 , wherein the step of predicting the second group's response comprises using a discriminative classifier to predict a crowd-sourced rating of the stimulus.

5. The method of claim 2 , wherein the group behavior comprises a collective coherency measurement corresponding to an engaged stimulus segment.

6. The method of claim 1 , wherein the step of generating an aggregate response to the stimulus comprises detecting a set of change-points for the response of each member of the group for the segment.

7. The method of claim 1 , wherein the step of generating the group behavior further comprises predicting a second group's response to the stimulus based on the entropy.

8. The method of claim 1 , wherein the period exceeds ten minutes.

9. The method of claim 1 , wherein the signal is a uniform visible signal.

10. The method of claim 9 , wherein the uniform visible signal is an output of an infra-red (IR) bandpass filter.

11. The method of claim 1 , wherein the step of detecting a response of each member comprises:
detecting a face of each member;
registering the face of each member;
detecting a facial motion of each member; and
recognizing the facial motion of each member by comparing the facial motion detected to a set of facial motion templates.

12. The method of claim 11 , wherein the step of aggregating the response of each member comprises detecting a synchrony of the facial motion among all members of the group.

13. The method of claim 1 , wherein the step of detecting a response of each member comprises creating a body motion profile for each member comprising:
extracting and registering each part of a body of each member;
measuring a movement of each part of the body by measuring a three-dimensional (3D) displacement; and
calculating a normalized motion for each part of the body.

14. The method of claim 13 , wherein the step of measuring the movement of each part comprises:
initializing a local temporal 3D volume for each member;
generating a flow-profile of each member contained within the corresponding temporal 3D volume; and
calculating a normalized local 3D energy for each member.

15. The method of claim 14 , wherein the step of aggregating the response of each member comprises determining a normalized 3D energy for the group based on the normalized local 3D energy for each member of the group.

16. The method of claim 1 , further comprising associating the group behavior with the stimulus.

17. The method of claim 1 , further comprising detecting a collective change in the group behavior in response to a change of an input feature of the stimulus.

18. The method of claim 1 , wherein all members of the group are exposed to the stimulus in an environment.

19. The method of claim 18 , wherein a first member is exposed to the stimulus at a first time and a second member is exposed to the stimulus at a second time, and the first time is different from the second time.

20. The method of claim 1 , wherein all members of the group are exposed to the stimulus at the same time, a first member of the group is in a first environment and a second member of the group is in a second environment when being exposed to the stimulus, the first environment is different from the second environment.

21. A system of measuring group behavior, comprising:
non-transitory memory storing a set of instructions; and
a processor coupled to the non-transitory memory, wherein the set of instructions are configured to cause the processor to:
obtain a signal comprising a sequence of images of a group having a plurality of members, wherein the group is exposed to a stimulus;
detect a response of each member of the group to the stimulus over a period; and
generate an aggregate response in response to the stimulus by aggregating the response of each member of the group; and
generate a group behavior measurement by determining an amount of synchrony of the response of each member relative to the aggregate response, wherein the step of generating the group behavior comprises:
generating a similarity matrix by determining a pair-wise correlation between two members of the group for a segment of the period; and
determining an entropy of the similarity matrix to measure the group behavior.

22. The system of claim 21 , wherein the signal is a uniform visible signal.

23. The system of claim 22 , wherein the uniform visible signal is an output of an infra-red (IR) bandpass filter.

24. The system of claim 21 , wherein the step of detecting a response of each member comprises:
detecting a face of each member;
registering the face of each member;
detecting a facial motion of each member; and
recognizing the facial motion of each member by comparing the facial motion detected to a set of facial motion templates.

25. The system of claim 24 , wherein the step of aggregating the response of each member comprises detecting a synchrony of the facial motion among all members of the group.

26. The system of claim 21 , wherein the step of detecting a response of each member comprises creating a flow profile for each member comprising:
initializing a local 3D volume for each member;
generating a flow-profile of each member contained within the corresponding 3D volume; and
calculating a normalized local 3D energy for each member.

27. The system of claim 26 , wherein the step of aggregating the response of each member comprises determining a normalized 3D energy for the group based on the normalized local 3D energy for each member of the group.

28. A method of measuring group behavior, comprising:
obtaining a signal comprising a sequence of images of a group having a plurality of members, wherein the group is exposed to a stimulus;
detecting a response of each member of the group to the stimulus over a period;
generating an aggregate response to the stimulus by aggregating the response of each member of the group; and
generating a group behavior measurement by determining an amount of synchrony of the response of each member relative to the aggregate response, wherein the step of generating the group behavior comprises:
segregating the response of each member over the period into a plurality of responses over a first set of time intervals, each time interval of the first set of time intervals compri

Assignees

• Disney Entpr Inc

Inventors

• Lucey Patrick
• Navarathna Rajitha
• Carr Peter
• Candela Daniel
• Matthews Iain

Classifications

• G06T2207/30201

  Face · CPC title

• G06K9/00302

  Physics · mapped topic

• G06K9/00335

  Physics · mapped topic

• G06T7/2033

  Physics · mapped topic

• G06T2207/10016

  Video; Image sequence · CPC title