Feature-Based Candidate Selection

• US 20150161126A1
• Filed: 08/30/2012
• Published: 06/11/2015
• Est. Priority Date: 08/30/2012
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Wang, Xin-Jing, Scott, Matthew Robert, Chen, Xi
• Granted Patent: US 9,767,156 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/248 Presentation of query results
• CPC Class Codes: G06F 16/9038 Presentation of query results
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 3/04842 Selection of displayed objects or displayed text elements G06F3/0482 takes precedence
• CPC Class Codes: G06F 40/274 Converting codes to words; Guess-ahead of partial word inputs

Abstract

A candidate for selection or insertion may be based on extracted candidate features and reference features. The candidate features may be scored and ranked while candidates may be removed from consideration based on criteria. The remaining candidates may be presented for selection and/or insertion.

Claims

1. A computer-readable storage media storing instructions that are executable by one or more processors to configure the one or more processors to:

select one or more references from one or more candidates returned by a search engine based at least in part on search criteria and results information, the search criteria including at least user input;

assign corresponding scores to the one or more candidates based at least in part on a comparison between one or more candidate features extracted from the one or more candidates and one or more reference features extracted from the one or more references;

select one or more of the candidates based at least in part on the score; and

provide the selected one or more of the one or more candidates for user selection.

2. The computer-readable storage media of claim 1, wherein the instructions are further executable to cause the processors to:

receive the user input comprising one or more of a textual input, a gesture, a motion, or a voice input;

analyze one or more of a context of the user input or a content of the user input; and

gather the one or more candidates based at least in part on the one or more of the context of the user input or the content of the user input.

3. The computer-readable storage media of claim 2, wherein the context of the user input comprises one or more of a previous search, a previous user input, or a computer program.

4. The computer-readable storage media of claim 2, wherein the one or more candidates gathered by the search engine are returned in a ranked order.

5. The computer-readable storage media of claim 4, wherein the results information is based at least in part on one or more of a set of top ranked results, a set of tail ranked results, or a set duplicate results.

6. The computer-readable storage media of claim 1, wherein the one or more candidates comprise one or more of a candidate image, a candidate video, candidate audio file, or a textualized graphic.

7. The computer-readable storage media of claim 1, wherein the reference features comprise average features of the one or more references.

8. The computer-readable storage media of claim 1, wherein the score is assigned based at least in part on a classifier.

9. The computer-readable storage media of claim 8, wherein the classifier assigns scores based at least in part on a learned model that maps features to a score value.

10. The computer-readable storage media of claim 8, wherein the classifier assigns scores based at least in part on a regression model.

11. The computer-readable storage media of claim 1, wherein the select one or more candidates based at least in part on the score comprises comparing the score to a threshold score value, wherein one or more candidates with scores greater than the threshold score value are available for selection.

12. The computer-readable storage media of claim 1, wherein the select one or more candidates based at least in part on the score comprises a comparison between a candidate and another candidate, wherein a candidate duplicative or within a similarity threshold of another candidate is made unavailable for selection.

13. The computer-readable storage media of claim 12, wherein the comparison of the candidate to the another candidate is based at least in part on one or more of local features, global features, or pixel-level comparisons.

14. A system comprising:

one or more processors;

an extractor executable by the one or more processors to extract one or more reference features from one or more reference images and to extract one or more candidate features from one or more candidate images, the one or more reference images selected, based at least in part on search results, from the one or more candidate images returned from a search engine;

a classifier executable by the one or more processors to assign a score to one or more candidate images based at least in part on a comparison between the extracted candidate features and the extracted reference features; and

an insertion manager executable by the one or more processors to provide the one or more candidate images for selection.

15. The system of claim 14, further comprising:

an analyzer executable by the one or more processors to analyze a context of the textual input, a content of the textual input, or a combination thereof; and

a candidate manager executable by the one or more processors to gather one or more candidate images based at least in part on the context of the textual input and the content of the textual input, and to select one or more reference images from the one or more candidate images based at least in part on one or more of a set of top ranked results, a set of tail ranked results, or a set duplicate results.

16. The system of claim 14, further comprising a reducer executable by the one or more processors to remove a first candidate image from the one or more candidate images based at least in part on a score of a candidate compared to a threshold score value.

17. The system of claim 14, further comprising a duplicate remover executable by the one or more processors to remove a second candidate image from the one or more candidates based at least in part on one or more of a duplicity of a candidate to another candidate or a similarity of a candidate to another candidate.

18. A method comprising:

receiving a textual input from a user through an input method editor (IME) interface;

analyzing, by a computing device, a content of the textual input and a context of the textual input;

gathering one or more candidate images from a search engine based at least in part on the context of the textual input and the content of the textual input;

selecting one or more reference images from the one or more candidate images based at least in part on results from the search engine;

extracting one or more reference features from the one or more reference images;

extracting one or more candidate features from the one or more candidate images;

assigning a score to one or more candidate images based at least in part on the extracted candidate features;

ranking one or more candidate images based at least in part on the score;

removing a first candidate image from the one or more candidate images when the score of the first candidate image is below a threshold score;

removing a second candidate image from the one or more candidates based at least in part on a comparison of features between the second candidate and another candidate; and

providing a ranked subset of remaining candidate images of the one or more candidates, to the user, for selection through the IME interface.

19. The method of claim 18, wherein the results from the search engine comprise one or more of a set of top ranked results, a set of tail ranked results, or a set duplicate results.