Rebuilding images based on historical image data

Patent metadata

Field	Value
Publication number	US-10032301-B2
Application number	US-201815893769-A
Country	US
Kind code	B2
Filing date	Feb 12, 2018
Priority date	Oct 6, 2016
Publication date	Jul 24, 2018
Grant date	Jul 24, 2018

Abstract

A sub-image of data of a first full image may be selected. The sub-image of data may at least partially obscure an object within the first full image. A request to replace the sub-image of data may be transmitted over a network. The request may include transmitting the full image and transmitting metadata associated with the first full image to one or more of a server computing device. The server computing device may analyze a history of images and select one or more images of the history of images that match one or more attributes of the metadata. The server computing device may replace the sub-image of data using the one or more images to generate at least a second full image that includes the object, wherein the object is not obscured. The second full image may be received over the network.

First claim

What is claimed is:

1. A system for selecting replacement image data, the system comprising:
a memory; and
a processor configured to obtain instructions from the memory that cause the processor to perform a method comprising:
receiving, over a network and from a first computing device, a user request to replace a first sub-image of data of a first full image, a portion of the first full image being obscured by the first sub-image of data, the first full image being a digital photograph;
receiving, over the network and from the first computing device, the first full image and metadata associated with the first full image, the metadata including attributes of: a first geolocation identifier of where the first full image was taken, a first set of exposure values of the first full image, a first timestamp identifier of when the first full image was taken, a distance measurement value indicating a distance from the first computing device to an object within an environment that the first full image was taken in, an angle value indicating an angle with which the first full image was taken at, a bit value indicating bits per pixel of the first full image, a set of contrast values of the first full image, a digital zoom ratio value of the first full image, a setting value indicating whether a flash setting was on or off for a camera of the first computing device that captured the first full image, a pixel height value of the first full image, a pixel width value of the first full image, a depth value of the first full image, a model value indicating a camera make and model of the camera, a set of resolution values of the first full image, a focal length value of the first full image, and a color space value of the first full image;
matching the first geolocation identifier to a second geolocation identifier associated with a second set of full images, the second set of full images being taken in a same location as the first full image, the second set of full images being downloaded from a plurality of user mobile devices to a central datastore;
matching the first timestamp identifier to a second timestamp identifier associated with a subset of the second set of full images, wherein the subset of the second set of full images is taken within a threshold time period of when the first full image was taken, the subset of the second set of full images matching the portion of the first full image in an unobscured state;
removing the first sub-image of data from the first full image;
replacing the first sub-image of data with a second sub-image of data based on the matching of the first geolocation identifier to the second geolocation identifier, and further based on the matching of the first time stamp identifier to the second timestamp identifier, wherein the second sub-image of data is included within one of the subset of the second set of full images; and
enhancing the second sub-image of data based on changing one or more features of the second sub-image to match the metadata associated with the first full image, the enhancing including adjusting a second set of exposure values of the second sub-image of data to match the first set of exposure values of the first full image.

Assignees

• IBM

Inventors

• Balasubramanian Swaminathan
• De Radha M
• Delport Ashley D
• Poddar Indrajit
• Vasudevan Cheranellore

Classifications

• G06F17/30867

  Physics · mapped topic

• G06F3/04842

  Selection of displayed objects or displayed text elements (G06F3/0482 takes precedence) · CPC title

• G06F17/30256

  Physics · mapped topic

• G06T11/60 Primary

  Creating or editing images; Combining images with text · CPC title

• G06F17/30268

  Physics · mapped topic