AUTOMATICALLY DETERMINING FIELD OF VIEW OVERLAP AMONG MULTIPLE CAMERAS

• US 20150379729A1
• Filed: 09/14/2015
• Published: 12/31/2015
• Est. Priority Date: 03/28/2013
• Status: Active Grant

• Current Assignee: Kyndryl Incorporated (Kyndryl Holdings, Inc.)
• Original Assignee: International Business Machines Corporation
• Inventors: Datta, Ankur, Feris, Rogerio S., Pankanti, Sharathchandra U., Zhai, Yun
• Granted Patent: US 9,710,924 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 2207/30232 Surveillance
• CPC Class Codes: G06T 2207/30252 Vehicle exterior; Vicinity of vehicle
• CPC Class Codes: G06T 7/251 involving models
• CPC Class Codes: G06T 7/292 Multi-camera tracking

Abstract

Field of view overlap among multiple cameras are automatically determined as a function of the temporal overlap of object tracks determined within their fields-of-view. Object tracks with the highest similarity value are assigned into pairs, and portions of the assigned object track pairs having a temporally overlapping period of time are determined. Scene entry points are determined from object locations on the tracks at a beginning of the temporally overlapping period of time, and scene exit points from object locations at an ending of the temporally overlapping period of time. Boundary lines for the overlapping fields-of-view portions within the corresponding camera fields-of-view are defined as a function of the determined entry and exit points in their respective fields-of-view.

Claims

1. A computer implemented method for automatically determining field of view overlap among multiple cameras as a function of the temporal overlap of object tracks, the method comprising executing on a processing unit the steps of:

determining values of similarity for potential pairings of each of a plurality of object tracks of movements of objects over time through a field-of-view of a first camera with each of a plurality of object tracks of movements of objects over time through a field-of-view of a second camera;

assigning into pairs each of the object tracks within the first camera field-of-view to ones each of the object tracks within the second camera field-of-view that each have a highest determined similarity value of the similarity values determined for their respective potential pairings; and

determining that the assigned object track pairs correspond to overlapping portions of the fields-of-view of the first and second paired cameras if portions of each of the assigned object track pairs have a temporally overlapping period of time.

2. The method of claim 1, further comprising:

integrating computer-readable program code into a computer system comprising the processing unit, a computer readable memory and a computer readable tangible storage medium; and

wherein the computer readable program code is embodied on the computer readable tangible storage medium and comprises instructions for execution by the processing unit via the computer readable memory to cause the processing unit to perform the steps of determining the value of similarity for the potential pairings of the object tracks, assigning into pairs the object tracks having the highest determined similarity values, and determining that the assigned object track pairs correspond to overlapping portions of the fields-of-view if portions of each have the temporally overlapping period of time.

3. The method of claim 1, further comprising:

determining the values of similarity for the potential pairings as a function of appearance information data of the respective objects or their tracks based on a strength of similarity of object features that include at least one of color, size and texture features.

4. The method of claim 3, wherein the step of assigning into the pairs each of the object tracks that have the highest determined similarity values for their respective potential pairings comprises:

matching objects as a function of a high-dimensional feature vector that comprises a plurality of histograms of color channels and oriented gradients texture features.

5. The method of claim 4, wherein the step of assigning into the pairs each of the object tracks that have the highest determined similarity values for their respective potential pairings comprises:

building a similarity matrix for different respective values of similarity comprising a plurality of locations that are each defined as pairings of one each of the object tracks in the first camera view to one each of the object tracks in the second camera view;

populating each of the matrix locations with similarity values of their respective track pairings;

discarding matrix locations populated with similarity values that have similarity values below a requisite minimum threshold value; and

determining the potential pairings as the object tracks corresponding to the populated matrix locations having the highest similarity values.

6. The method of claim 5, further comprising:

determining scene entry points on each of the tracks of the assigned object track pairs corresponding to the overlapping fields-of-view portions within their fields-of-view as locations of the object on the each of said tracks at a beginning of the temporally overlapping period of time;

determining scene exit points on each of tracks of the assigned object track pairs corresponding to the overlapping fields-of-view portions within their fields-of-view as locations of the object on the each of said tracks at an ending of the temporally overlapping period of time; and

defining boundary lines for the overlapping fields-of-view portions within their corresponding first and second camera fields-of-view as a function of the determined entry and exit points in their respective fields-of-view.

7. The method of claim 6, further comprising:

readjusting a first feature weighting used to match appearances of objects represented by respective tracks of the assigned object tracks downward relative to a weighting of a second feature used to match the appearances of objects in response to;

the first feature indicating a first match of objects between the first and second camera fields-of-view that is located outside of the overlapping fields-of-view areas defined by the boundary lines; and

the second feature indicating a second match of objects between the first and second camera fields-of-view that is located within the overlapping fields-of-view areas defined by the boundary lines.

8. The method of claim 6, wherein the step of defining the boundary lines comprises determining linear boundary lines through the fields of view that comprise subsets of the determined entry and exit points in the respective fields-of-view.

9. The method of claim 6, wherein the step of defining the boundary lines comprises determining linear solutions using robust regression through the subsets of the determined entry and exit points in the respective fields-of-view.

10. The method of claim 6, further comprising:

in response to a determined change in a relative difference in motion or size of a matched appearance of an object represented by respective tracks of one of the assigned object track pairs corresponding to the overlapping fields-of-view portions over time, determining a displacement distance between before and after locations of the object in one of the fields of views of the one assigned object track pair; and

updating a location of at least one of the boundary lines within the fields of view of the one assigned object track pair by the determined displacement distance.

11. A system, comprising:

a processing unit;

a computer readable memory in communication with the processing unit; and

a computer-readable storage medium in communication with the processing unit;

wherein the processing unit executes program instructions stored on the computer-readable storage medium via the computer readable memory and thereby;

determines values of similarity for potential pairings of each of a plurality of object tracks of movements of objects over time through a field-of-view of a first camera with each of a plurality of object tracks of movements of objects over time through a field-of-view of a second camera;

assigns into pairs each of the object tracks within the first camera field-of-view to ones each of the object tracks within the second camera field-of-view that each have a highest determined similarity value of the similarity values determined for their respective potential pairings; and

determines that the assigned object track pairs correspond to overlapping portions of the fields-of-view of the first and second paired cameras if portions of each of the assigned object track pairs have a temporally overlapping period of time.

12. The system of claim 11, wherein the processing unit executes the program instructions stored on the computer-readable storage medium via the computer readable memory and thereby determines the values of similarity for the potential pairings as a function of appearance information data of the respective objects or their tracks based on a strength of similarity of object features that include at least one of color, size and texture features.

13. The system of claim 11, wherein the processing unit executes the program instructions stored on the computer-readable storage medium via the computer readable memory and assigns into the pairs each of the object tracks that have the highest determined similarity values for their respective potential pairings by matching objects as a function of a high-dimensional feature vector that comprises a plurality of histograms of color channels and oriented gradients texture features.

14. The system of claim 13, wherein the processing unit executes the program instructions stored on the computer-readable storage medium via the computer readable memory and assigns into the pairs each of the object tracks that have the highest determined similarity values for their respective potential pairings by:

building a similarity matrix for different respective values of similarity comprising a plurality of locations that are each defined as pairings of one each of the object tracks in the first camera view to one each of the object tracks in the second camera view;

populating each of the matrix locations with similarity values of their respective track pairings;

discarding matrix locations populated with similarity values that have similarity values below a requisite minimum threshold value; and

determining the potential pairings as the object tracks corresponding to the populated matrix locations having the highest similarity values.

15. The system of claim 14, wherein the processing unit executes the program instructions stored on the computer-readable storage medium via the computer readable memory and thereby:

determines scene entry points on each of the tracks of the assigned object track pairs corresponding to the overlapping fields-of-view portions within their fields-of-view as locations of the object on the each of said tracks at a beginning of the temporally overlapping period of time;

determines scene exit points on each of tracks of the assigned object track pairs corresponding to the overlapping fields-of-view portions within their fields-of-view as locations of the object on the each of said tracks at an ending of the temporally overlapping period of time; and

defines boundary lines for the overlapping fields-of-view portions within their corresponding first and second camera fields-of-view as a function of the determined entry and exit points in their respective fields-of-view.

16. A computer program product for automatically determining field of view overlap among multiple cameras as a function of the temporal overlap of object tracks, the computer program product comprising:

a computer readable hardware storage device having computer readable program code embodied therewith, wherein the computer readable hardware storage device is not a transitory signal per se, the computer readable program code comprising instructions for execution by a computer processing unit that cause the computer processing unit to;

determine values of similarity for potential pairings of each of a plurality of object tracks of movements of objects over time through a field-of-view of a first camera with each of a plurality of object tracks of movements of objects over time through a field-of-view of a second camera;

assign into pairs each of the object tracks within the first camera field-of-view to ones each of the object tracks within the second camera field-of-view that each have a highest determined similarity value of the similarity values determined for their respective potential pairings; and

determine that the assigned object track pairs correspond to overlapping portions of the fields-of-view of the first and second paired cameras if portions of each of the assigned object track pairs have a temporally overlapping period of time.

17. The computer program product of claim 16, wherein the computer readable program code instructions for execution by the computer processing unit further cause the computer processing unit to determine the values of similarity for the potential pairings as a function of appearance information data of the respective objects or their tracks based on a strength of similarity of object features that include at least one of color, size and texture features.

18. The computer program product of claim 16, wherein the computer readable program code instructions for execution by the computer processing unit further cause the computer processing unit to assign into the pairs each of the object tracks that have the highest determined similarity values for their respective potential pairings by matching objects as a function of a high-dimensional feature vector that comprises a plurality of histograms of color channels and oriented gradients texture features.

19. The computer program product of claim 18, wherein the computer readable program code instructions for execution by the computer processing unit further cause the computer processing unit to assign into the pairs each of the object tracks that have the highest determined similarity values for their respective potential pairings by:

building a similarity matrix for different respective values of similarity comprising a plurality of locations that are each defined as pairings of one each of the object tracks in the first camera view to one each of the object tracks in the second camera view;

populating each of the matrix locations with similarity values of their respective track pairings;

discarding matrix locations populated with similarity values that have similarity values below a requisite minimum threshold value; and

determining the potential pairings as the object tracks corresponding to the populated matrix locations having the highest similarity values.

20. The computer program product of claim 19, wherein the computer readable program code instructions for execution by the computer processing unit further cause the computer processing unit to:

determine scene entry points on each of the tracks of the assigned object track pairs corresponding to the overlapping fields-of-view portions within their fields-of-view as locations of the object on the each of said tracks at a beginning of the temporally overlapping period of time;

determine scene exit points on each of tracks of the assigned object track pairs corresponding to the overlapping fields-of-view portions within their fields-of-view as locations of the object on the each of said tracks at an ending of the temporally overlapping period of time; and

define boundary lines for the overlapping fields-of-view portions within their corresponding first and second camera fields-of-view as a function of the determined entry and exit points in their respective fields-of-view.