Agent matching based on video analysis of customer presentation

Patent metadata

Field	Value
Publication number	US-9451087-B2
Application number	US-201213447943-A
Country	US
Kind code	B2
Filing date	Apr 16, 2012
Priority date	Apr 16, 2012
Publication date	Sep 20, 2016
Grant date	Sep 20, 2016

Abstract

Systems and methods for routing and/or servicing contacts using video analysis of one or more video streams are provided. The systems and methods are particularly applicable to a contact center.

First claim

What is claimed is:

1. A method comprising:
segmenting, by a processor executing a video analysis module, a video stream to produce segmented video analysis results, the segmenting using background and foreground models retrieved from one or more databases and comprising detecting a change, measuring a degree of the change, localizing the change, and extracting relevant changes;
classifying, by the processor executing the video analysis module, the segmented video analysis results using one or more appearance models retrieved from one or more databases to generate classified video analysis results;
tracking, by the processor executing the video analysis module, the classified video analysis results using one or more motion and camera models retrieved from one or more databases to generate tracked video analysis results;
recognizing, by the processor executing the video analysis module, at least one object present in the video stream to generate an activity report, the recognizing comprising combining the classified video analysis results and the tracked video analysis results; and
selecting, by the processor executing a work assignment engine and based on the activity report, a communication device of a resource from among a plurality of communication devices of a plurality of resources to initially service a contact of a contact center.

2. The method of claim 1 , wherein a source of the video stream is the selected communication device and wherein the video stream comprises an image of the resource.

3. The method of claim 2 , wherein the recognizing at least one object present in the video stream further comprises analyzing the at least one object to automatically determine at least one sales opportunity; and
comparing at least a portion of the activity report with video analysis results from analysis of a second video stream comprising an image of a contactor associated with the contact.

4. The method of claim 1 , wherein a source of the video stream is a communication device of a contactor associated with the contact, the video stream comprises an image of the contactor, and the activity report comprises an identification instance of one or more of the following: a bodily movement of the contactor, a bodily position of the contactor, a bodily orientation of the contactor, an item or article of clothing of the contactor, a gesture of the contactor, a mannerism of the contactor, a medical condition of the contactor, and a non-human object in a scene provided by the video stream.

5. The method of claim 4 , wherein the activity report comprises a prediction of one or more of the following: a potential sale or upsell opportunity, an identity of the contactor, a socioeconomic class of the contactor, a customer class of the contactor, a geographical location of the contactor, an age of the contactor, a profession of the contactor, a need of the contactor, a manufacture, brand or model of an item associated with the contactor, and a purpose of the contact.

6. The method of claim 4 , wherein the resource is an emergency responder and the contact center is a Public Safety Access Point contact center.

7. The method of claim 6 , wherein the activity report is used for one or more of: providing advice to the emergency responder, selecting a type of emergency responder to be dispatched, determining a geographic location of the contactor, determining an urgency in dispatching an emergency responder, and identifying contacts associated with a common emergency event.

8. A work assignment system comprising:
a processor; and
a computer readable medium coupled to the processor and comprising instructions that, when executed by the processor, cause the processor to:
segment a video stream to produce segmented video analysis results, the segmenting using background and foreground models retrieved from one or more databases and comprising detecting a change, measuring a degree of the change, localizing the change, and extracting relevant changes;
classify the segmented video analysis results using one or more appearance models retrieved from one or more databases to generate classified video analysis results;
track the classified video analysis results using one or more motion and camera models retrieved from one or more databases to generate tracked video analysis results;
recognize at least one object present in the video stream to generate an activity report, the recognizing comprising combining the classified video analysis results and the tracked video analysis results; and
select, based on the activity report, a communication device of a resource from among a plurality of communication devices of a plurality of resources to initially service a contact of a contact center.

9. The work assignment system of claim 8 , wherein a source of the video stream is the selected communication device and wherein the video stream comprises an image of the resource.

10. The work assignment system of claim 9 , wherein the instructions, when executed by the processor, further cause the processor to:
compare at least a portion of the activity report with video analysis results from analysis of a second video stream comprising an image of a contactor associated with the contact.

11. The work assignment system of claim 8 , wherein a source of the video stream is a contactor associated with the contact, the video stream comprises an image of the contactor, and the activity report comprises an identification instance of one or more of the following: a bodily movement of the contactor, a bodily position of the contactor, a bodily orientation of the contactor, an item or article of clothing of the contactor, a gesture of the contactor, a mannerism of the contactor, a medical condition of the contactor, and a non-human object in a scene provided by the video stream.

12. The work assignment system of claim 11 , wherein recognizing further comprises analyzing the at least one object to automatically determine at least one sales opportunity, and the activity report comprises a prediction of one or more of the following: a potential sale or upsell opportunity, an identity of the contactor, a socioeconomic class of the contactor, a customer class of the contactor, a geographical location of the contactor, an age of the contactor, a profession of the contactor, a need of the contactor, a manufacture, brand or model of an item associated with the contactor, and a purpose of the contact.

13. The work assignment system of claim 8 , wherein the resource is an emergency responder, the contact center is a Public Safety Access Point contact center, and the activity report comprises one or more of: advice for the emergency responder, identification of a type of emergency responder to be dispatched, information corresponding to a geographic location of the contactor, information corresponding to an urgency in dispatching an emergency responder, and identification of contacts associated with a common emergency event.

14. A contact center, comprising:
a processor; and
a computer readable medium, coupled to the processor, comprising instructions that program the processor to:
segment a video stream using background and foreground models retrieved from one or more databases to produce segmented video analysis results, wherein the processor detects a change, measures a degree of the change, localizes the change, and extracts relevant changes;
classify the segmented video analysis results using one or more appearance models retrieved from one or more databases to produce classified video analysis results,
track the classified video analysis results using one or more motion and camera mode

Assignees

• Fagundes Luciano Godoy
• Moran Thomas
• Desai Dhaval
• Kohler Joylee
• Michaelis Paul Roller
• Avaya Inc

Inventors

• Fagundes Luciano Godoy
• Moran Thomas
• Desai Dhaval
• Kohler Joylee
• Michaelis Paul Roller

Classifications

• H04M2201/50

  Telephonic communication in combination with video communication · CPC title

• H04M2201/18

  Comparators · CPC title

• H04M3/523 Primary

  with call distribution or queueing · CPC title

• H04M3/5116

  for emergency applications · CPC title

• H04M3/51

  Centralised call answering arrangements requiring operator intervention {, e.g. call or contact centers for telemarketing} · CPC title