IP Library › Granted Patent US 10,409,854
Granted Patent B2
US 10,409,854 · App. 15/545,812 · Granted Sep 10, 2019

Image selection based on text topic and image explanatory value

Inventors: Lei Liu (Palo Alto, CA); Jerry Liu (Palo Alto, CA); Shanchan Wu (Palo Alto, CA); Hector A Lopez (Palo Alto, CA)
Assignee: Hewlett-Packard Development Company, L.P.
G06F16/5838G06F16/951G06F17/218G06F17/241G06F17/2785G06K9/00456
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 10,409,854
App. No.
15/545,812
Granted
Sep 10, 2019
Kind
B2
Abstract

Examples disclosed herein relate to selecting an image based on text topic and image explanatory value. In one implementation, a processor selects an image to associate with a text based on a criteria indicating the explanatory value of a context information related to the image in relation to the topic of the text. The processor may output the selected image.

Claims (38)

1. A computing system, comprising:

a processor to:

determine a topic associated with a set of text;

generate a query based on the topic, wherein query comprises a question, wherein a set of images is found based on the query, wherein the set of images comprises text associated with a respective image that answers the question;

select a top N images within the set of images based on a level of explanatory value of content associated with the top N images compared to the topic that is determined, wherein the level of explanatory value of the top N images is above a threshold;

provide an icon by text associated with the top N images to indicate that the top N images are available for the text, wherein a user interface displays the top N images to allow a user to select an image from the top N images from the user interface; and

output information related to the image that is selected.

2. The computing system of claim 1 , wherein to select the top N images within the set of images based on the level of explanatory value comprises the processor to:

determine an ability of the top N images to address the question based on the content associated with the top N images; and

select the top N images based on the ability of the top N images to address the question.

3. The computing system of claim 1 , wherein to select the top N images within the set of images based on the level of explanatory value comprises the processor to:

identify knowledge phrases in the content associated with the top N images; and

select the top N images based on the knowledge phrases that are identified.

4. The computing system of claim 1 , wherein the processor is further to select the set of images based on relevance to the topic.

5. The computing system of claim 1 , wherein to select the top N images based on the level of explanatory value comprises the processor to:

determine a topic of the content associated with the top N images; and

select the top N images based on a comparison of the topic associated with the set of text that is determined and the topic of the content associated with the top N images that is determined.

6. The computing system of claim 1 , wherein to determine the topic associated with the set of text comprises the processor to:

determine a topic associated with a portion of the set of text; and

divide the set of text into portions based on a difference in the topic associated with the portions.

7. The computing system of claim 6 , wherein to divide the set of text into the portions comprises the processor to combine adjacent portions based on a similarity of topics.

8. A method, comprising:

determining, by a processor, a topic associated with a text;

generating a query based on the topic, wherein query comprises a question, wherein a set of images is found based on the query, wherein the set of images comprises text associated with a respective image that answers the question;

selecting a set of images based on the topic that is determined;

selecting a top N images within the set of images based on an explanatory value of the top N images compared to the topic associated with the text, wherein the explanatory value is related to a comparison of a topic associated with the top N images and the topic associated with the text, wherein the explanatory value of the top N images is above a threshold;

providing an icon by text associated with the top N images to indicate that the top N images are available for the text, wherein a user interface displays the top N images to allow a user to select an image from the top N images from the user interface; and

associating the image that is selected with the text based on a ranking.

9. The method of claim 8 , wherein the selecting the set of top N images comprises applying the search query to a web based search engine.

10. The method of claim 8 , further comprising filtering the set of images based on at least one of: a difficulty level, display characteristics, and permissions settings.

11. A machine-readable non-transitory storage medium comprising instructions executable by a processor to:

generate a query based on a topic, wherein query comprises a question, wherein a set of images is found based on the query, wherein the set of images comprise text associated with a respective image that answers the question;

select a top N images from the set of images to associate with a text based on a criteria indicating an explanatory value of a context information related to the top N images in relation to the topic of the text, wherein the explanatory value of the top N images is above a threshold;

provide an icon by text associated with the top N images to indicate that the top N images are available for the text, wherein a user interface displays the top N images to allow a user to select an image from the top N images from the user interface; and

output the image that is selected.

12. The machine-readable non-transitory storage medium of claim 11 , wherein the instructions to select the top N images comprise instructions to select the top N images based on a comparison of a semantic topic associated with the text and a semantic topic associated with a document including the top N images.

13. The machine-readable non-transitory storage medium of claim 11 , further comprising instructions to divide the text into portions based on divisions in the topic, wherein the top N images that is selected is selected for a portion of the text.

14. The machine-readable non-transitory storage medium of claim 12 , wherein instructions to output the image comprise instructions to create a publication including the text and the image.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jun 19, 2018
From: LIU, LEI; LIU, JERRY J.; WU, SHANCHAN; ARMANDO LOPEZ, HECTOR
To: HEWLETT-PACKARD DEVELOPMENT COMPANY, L.P.
Reel/Frame 046130/0430 →
Continuity (1)
Related Publication 20180018349A1 · Jan 18, 2018