Training image adjustment preferences

• US 9,412,046 B2
• Filed: 10/10/2014
• Issued: 08/09/2016
• Est. Priority Date: 10/10/2014
• Status: Active Grant

• Current Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Original Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Inventors: Shen, Shizhe, Maguire, Yael, Kowalewski, Damian, Passichenko, Viktor Vladimirovich
• Primary Examiner(s): Kassa, Yosef
• Application Number:
  US14/511,597
• Publication Number:
  US 20160104057A1
• Time in Patent Office: 669 Days
• Field of Search: 382/156, 382/158, 382/159, 382/209, 382/218, 382/278, 358/537, 358/452
• US Class Current: 1/1
• CPC Class Codes: G06F 16/583 using metadata automatically derived from the content
• CPC Class Codes: G06F 18/214 Generating training patterns; Bootstrap methods, e.g. bagging or boosting
• CPC Class Codes: H04N 1/00204 with a digital computer or a digital computer system, e.g. an internet server programmed control between transmitter and receiver or between image input and image output device H04N1/32561
• CPC Class Codes: H04N 2201/0084 Digital still camera
• CPC Class Codes: H04N 23/57 Mechanical or electrical details of cameras or camera modules specially adapted for being embedded in other devices

Abstract

Some embodiments include a method of operating a computing device to learn user preferences of how to process digital images. The method can include: aggregating a user image selection and a context attribute associated therewith into a preference training database for a user, wherein the user image selection represents a record of the user'"'"'s preference over at least one of adjusted versions of a base image when the adjusted versions are separately processed by different visual effects; determining a visual effect preference associated based on machine learning or statistical analysis of user image selections in the preference training database, the user image selections representing experimental records corresponding to the visual effects; updating a photo preference profile with the visual effect preference; and providing the photo preference profile to an image processor to adjust subsequently captured photographs provided to the image processor.

Claims

1. A computer-implemented method comprising:

aggregating, by a computing device, a first user image selection and a context attribute associated with the first user image selection into a preference training database for a user, wherein the first user image selection represents a record of how the user has preferred over at least one of adjusted versions of a base image when the adjusted versions are separately processed by visual effects that are different;

determining, by the computing device, a visual effect preference associated with the user based on machine learning or statistical analysis of user image selections in the preference training database, wherein the user image selections represent experimental records corresponding to the visual effects, wherein the visual effect preference is an image processing rule that is particular to the context attribute, and wherein the image processing rule specifies a visual effect process to execute when a digital image is determined to be associated with the context attribute;

updating, by the computing device, a photo preference profile with the visual effect preference; and

providing, by the computing device, the photo preference profile to an image processor to adjust subsequently captured photographs provided to the image processor.

2. The computer-implemented method of claim 1, wherein the computing device is a computer server connected via a network connection to a camera-enabled device associated with the user, the camera-enabled device having generated the base image.

3. The computer-implemented method of claim 2, wherein the image processor is part of the camera-enabled device.

4. The computer-implemented method of claim 1, wherein the computing device is a camera-enabled device associated with the user, the camera-enabled device having captured the base image.

5. The computer-implemented method of claim 1, wherein determining the visual effect preference includes performing the machine learning or statistical analysis on the user image selections associated with the context attribute.

6. The computer-implemented method of claim 1, wherein the context attribute is a camera module configuration, an image analysis context, a camera operation context, an environment context, or any combination thereof.

7. The computer-implemented method of claim 1, wherein the context attribute is a display device context associated with a display used to present the adjusted versions.

8. The computer-implemented method of claim 1, wherein the visual effects are image filters of the same type.

9. The computer-implemented method of claim 1, further comprising the image processor adjusting digital images that are owned by the user according to the photo preference profile of the user.

10. The computer-implemented method of claim 1, further comprising the image processor adjusting digital images that are being viewed by the user according to the photo preference profile of the user.

11. The computer-implemented method of claim 1, wherein providing the photo preference profile to the image processor is to adjust existing images in a memory device that is accessible to the image processor.

12. The computer-implemented method of claim 1, further comprising computing a group photo preference profile by averaging photo preference profiles from a group of multiple users, the photo preference profiles including the photo preference profile of the user.

13. A computer-readable storage memory storing computer-executable instructions comprising:

instructions for generating at least two digital image versions based on a digital image by applying at least two visual effects that are different;

instructions for causing the digital image versions to be displayed at a display device;

instructions for recording a user image selection from amongst the digital image versions, in response to displaying the digital image versions;

instructions for identifying a context attribute to associate with the digital image; and

instructions for providing the user image selection associated with the context attribute to a machine learning engine to update a photo preference profile associated a user, wherein updating the photo preference profile includes determining an image processing rule particular to the context attribute, and wherein the image processing rule specifies a visual effect process to execute when a digital image is determined to be associated with the context attribute.

14. The computer-readable storage memory of claim 13, wherein generating the at least two digital image versions is in response to capturing a photograph via a camera module.

15. The computer-readable storage memory of claim 13, further comprising:

instructions for identifying a display device specification of the display device used to present the digital image versions; and

instructions for providing the display device specification to the machine learning engine.

16. The computer-readable storage memory of claim 13, further comprising:

instructions for receiving the photo preference profile from the machine learning engine; and

instructions for configuring an image signal processor to apply a visual effect to digital images subsequently provided to the image signal processor according to the photo preference profile.

17. The computer-readable storage memory of claim 13, further comprising:

instructions for receiving the photo preference profile from the machine learning engine; and

instructions for configuring an image signal processor to apply a visual effect to existing digital images saved in a memory accessible by the image signal processor according to the photo preference profile.

18. The computer-readable storage memory of claim 13, further comprising:

instructions for implementing the machine learning engine to compute the photo preference profile by comparing multiple user image selections from multiple comparisons of adjusted image versions generated using the visual effects.

19. A computing device comprising:

a component for generating at least two digital image versions based on a digital image by applying at least two visual effects that are different;

a component for causing the digital image versions to be displayed at a display device;

a component for recording a first user image selection from amongst the digital image versions, in response to displaying the digital image versions;

a component for identifying a context attribute to associate with the digital image; and

a component for updating a photo preference profile associated with a user by applying machine learning or statistical analysis on multiple user image selections associated with the context attribute, the multiple user image selections including the first user image selection;

wherein updating the photo preference profile includes determining an image processing rule particular to the context attribute, and wherein the image processing rule specifies a visual effect process to execute when a digital image is determined to be associated with the context attribute.