Real-time retail marketing system and method

• US 7,240,834 B2
• Filed: 03/21/2005
• Issued: 07/10/2007
• Est. Priority Date: 03/21/2005
• Status: Expired due to Fees

• Current Assignee: Mitsubishi Electric Research Laboratories, Inc. (Mitsubishi Electric Corporation)
• Original Assignee: Mitsubishi Electric Research Laboratories, Inc. (Mitsubishi Electric Corporation)
• Inventors: Nikovski, Daniel N., Divakaran, Ajay, Kato, Mamoru
• Primary Examiner(s): Lee; Michael G.
• Assistant Examiner(s): FRANKLIN, JAMARA ALZAIDA
• Application Number:
  US11/085,323
• Publication Number:
  US 20060208070A1
• Time in Patent Office: 841 Days
• Field of Search: 235/383, 235/385, 705/14, 705/22, 705/28, 705/10
• US Class Current: 235/385
• CPC Class Codes: G06Q 20/203 Inventory monitoring
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0268 at point-of-sale [POS]
• CPC Class Codes: G06Q 30/0269 based on user profile or attribute

Abstract

A marketing system and method for a retail environment periodically reads RFID tags attached to products to produce a list of product identifications. Consumer recommendation rules are updated according to each list, and recommendations are generated according to the updated consumer recommendation rules. Then, content can be displayed in the retail environment based on the recommendations.

Claims

1. A marketing method for a retail environment, comprising:

reading periodically RFID tags attached to products in a retail environment to produce a plurality of lists of product identifications;

updating consumer recommendation rules according to changes in the product identifications in the plurality of lists;

generating recommendations according to the updated consumer recommendation rules;

sensing consumers near the products to generate trajectories of the consumers in the retail environment;

updating the consumer recommendation rules according to the trajectories; and

displaying content in the retail environment based on the recommendations.

2. The method of claim 1, in which the updating and generating uses a thin singular value decomposition.

3. The method of claim 1, further comprising:

associating a time with each product identification.

4. The method of claim 1, in which the content is predetermined when a current list is identical to a predetermined number of previous lists.

5. The method of claim 1, in which the sensing is performed by motion sensors.

6. The method of claim 1, further comprising:

acquiring a sequence of images of consumers in the retail environment;

generating trajectories of the consumers in the retail environment from the sequence of images; and

updating the consumer recommendation rules according to the trajectories.

7. The method of claim 6, in which consumer characteristics are determined from the sequence of images, and further comprising:

updating the consumer recommendation rules according to the consumer characteristics.

8. The method of claim 7, in which the consumer characteristics include height, gender, race, and approximate age.

9. The method of claim 1, further comprising:

acquiring a sequence of images of consumers in the retail environment;

recognizing faces in the sequence of images; and

updating the consumer recommendation rules according to the recognized faces.

10. The method of claim 1, in which the updating is incremental in real time.

11. The method of claim 1, in which the updating is periodic.

12. The method of claim 1, in which the sensing is active by use of an infrared proximity sensor.

13. The method of claim 1, in which the sensing is passive by use of an infrared motion sensor.

14. The method of claim 1, further comprising:

acquiring a sequence of images of the products and consumers in the retail environment;

generating trajectories of consumers in the retail environment from the sequence of images; and

updating the consumer recommendation rules according to the trajectories.

15. The method of claim 1, in which the updating and generating uses collaborative filtering.

16. The method of claim 1, in which the updating and generating uses an amount of time between changes on the plurality of lists for each product identification.

17. The method of claim 1, in which the updating and generating uses an amount of time that a current list is identical to a list.

18. The method of claim 1, in which the retail environment includes multiple stores and multiple display areas for arranging the products.

19. A marketing system for a retail environment, comprising:

a plurality of RFID tags attached to products in a retail environment;

an RFID reader configured to periodically read the RFID tags;

means for updating consumer recommendation rules according to changes in a plurality of lists of product identifications produced by the RFID reader;

means for generating recommendations according to the updated consumer recommendation rules;

a sensor configured to detect consumers near the products to generate trajectories of consumers in the retail environment;

means for updating the consumer recommendation rules according to the trajectories; and

a display device configured to display content in the retail environment based on the recommendations.

20. The system of claim 19, in which the sensing is performed by motion sensors.

21. The system of claim 19, further comprising:

a camera configured to acquire a sequence of images of consumers in the retail environment;

means for generate trajectories of consumers in the retail environment from the sequence of images; and

means for updating the consumer recommendation rules according to the trajectories.