IMAGE DIRECTED SEARCH

• US 20160283595A1
• Filed: 06/10/2016
• Published: 09/29/2016
• Est. Priority Date: 05/01/2013
• Status: Active Grant

• Current Assignee: Cloudsight Incorporated
• Original Assignee: CamFind Inc.
• Inventors: FOLKENS, BRADFORD A., MAZUR, DOMINIK K.
• Granted Patent: US 10,223,454 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/3334 Selection or weighting of terms from queries, including natural language queries
• CPC Class Codes: G06F 16/51 Indexing; Data structures therefor; Storage structures
• CPC Class Codes: G06F 16/583 using metadata automatically derived from the content
• CPC Class Codes: G06F 16/5866 using information manually generated, e.g. tags, keywords, comments, manually generated location and time information
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 3/013 Eye tracking input arrangements G06F3/015 takes precedence
• 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: G06F 3/04842 Selection of displayed objects or displayed text elements G06F3/0482 takes precedence
• CPC Class Codes: G06Q 30/0241 Advertisements
• CPC Class Codes: G06Q 30/0282 Rating or review of business operators or products
• CPC Class Codes: G06V 10/225 based on a marking or identifier characterising the area
• CPC Class Codes: G06V 10/235 based on user input or interaction
• CPC Class Codes: G06V 20/20 in augmented reality scenes

Abstract

An image recognition approach employs both computer generated and manual image reviews to generate image tags characterizing an image. The computer generated and manual image reviews can be performed sequentially or in parallel. The generated image tags may be provided to a requester in real-time, be used to select an advertisement, and/or be used as the basis of an internet search. In some embodiments generated image tags are used as a basis for an upgraded image review. A confidence of a computer generated image review may be used to determine whether or not to perform a manual image review. Generated image tags are optionally used for sorting specific domains such as the inventory of a vendor.

Claims

1. A search system comprising:

an I/O configured to communicate an image over a communication network;

image tagging logic configured to provide image tags characterizing the image, the tags including a plurality of tokens;

a token ranker configured to parse the image tags so as to identify the plurality of tokens and to rank the identified tokens;

a search engine configured to perform a search using the image tags;

response logic configured to provide results of the search, the results provided in an order responsive to the rank of the identified tokens; and

an electronic processor configured to execute at least the token ranker.

2. The system of claim 1, wherein the image tagging logic includes an automatic identification interface configured to communicate the image to an automatic identification system and to receive a computer generated review of the image from the automatic identification system, the computer generated review including the image tags.

3. The system of claim 2, wherein the image tagging logic includes destination logic configured to determine a first destination to send the image to, for a first manual review of the image by a first human reviewer, and includes image posting logic configured to post the image to the first destination.

4. The system of claim 1, wherein the image tagging logic is also configured to determine a domain of the image.

5. The system of claim 4, wherein the domain is determined based on the contents of the image.

6. The system of claim 4, wherein the domain is determined based on a source of the image.

7. The system of claim 4, wherein the search engine is configured to perform the search within the domain.

8. The system of claim 4, wherein the domain includes a specific inventory of products.

9. The system of claim 8, wherein the products in the specific inventory are each characterized by product tags and the search engine is configured to match the plurality of tokens to the product tags.

10. The system of claim 9, wherein the product tags are generated by tagging images of the products.

11. The system of claim 9, wherein the product tags are generated using a product name or product description.

12. The system of claim 9, wherein the product tags are weighted and the order in which the results of the search is provided is responsive to the weighting of the product tags.

13. The system of claim 1, wherein the token ranker is configured to assign values to the tokens and configured to use the assigned values to rank the identified tokens.

14. The system of claim 1, wherein the token ranker is configured to assign values to the tokens based on a uniqueness of each token.

15. The system of claim 1, wherein the token ranker is configured to identify phrases within the plurality of tokens and treat an identified phrase as a single token by assigning a single rank to the identified phrase.

16. A method of processing an image, the method comprising:

determining a weighting of image tokens;

receiving an image;

identifying image tags characterizing the image;

parsing the image tags to identify image tokens within the image tags;

performing a search using the tokens;

sorting results of the search based on the weighting of the image tokens; and

presenting the sorted results of the search.

17. The method of claim 16, wherein the weighting of age tokens is determined within a specific domain.

18. The method of claim 16, further comprising determining a weighing of phrases comprised of image tokens.

19. The method of claim 16, wherein the weighting of image tokens is based on a uniqueness of each image token.

20. The method of claim 16, wherein the weighting of image tokens is based on value of the each token within a product classification system

21. The method of claim 16, further comprising identifying synonyms of the tokens.

22. The method of claim 16, wherein the image is received from an application executed on a mobile device.

23. The method of claim 16, wherein the image tags are characterized using both an automatic identification system and a human image reviewer.

24. The method of claim 16, further including identifying a domain of the image.

25. The method of claim 24, wherein the domain is identified based on a source of the image.

26. The method of claim 24, wherein the domain is identified based on a one or more of the image tokens.

27. The method of claim 24, wherein the domain includes a specific product inventory.

28. The method of claim 24, wherein the search is performed on a database selected in response to the domain.

29. The method of claim 24, wherein the search is performed using the domain and the image tokens.

30. The method of claim 17, where the sorting is further based on product tags associated with products within the results, or on a calculation dependent on image token weight and/or product tag weight.

31. The method of claim 17, wherein the presented results include opportunities to purchase product or include reviews of products.