Image manipulation for electronic display

Patent metadata

Field	Value
Publication number	US-9684987-B1
Application number	US-201514632983-A
Country	US
Kind code	B1
Filing date	Feb 26, 2015
Priority date	Feb 26, 2015
Publication date	Jun 20, 2017
Grant date	Jun 20, 2017

Abstract

The density of images to display can be increased, and distractions reduced, through intelligent cropping or manipulation of at least some of the images. For objects such as dresses represented in the images, the density can be increased by cropping away regions of background outside the object region(s). Locating regions representing the face and legs of the wearer can enable cropping of the top and/or bottom of the image in order to cause the dress to occupy the majority of the area of the image, and can provide for a level of consistency of the sizes of the objects across the images, regardless of the sources of the images. Representative colors of the objects can also be selected to adjust the background color, in order to provide for easy distinction between the images while not providing contrasting or unappealing colors that take away from the aesthetics of the objects.

First claim

What is claimed is:

1. A system, comprising:
at least one processor; and
memory storing instructions that, when executed by the at least one processor, cause the system to:
obtain an image including a representation of an object, the object including an apparel item worn by a person, the image further including at least one region associated with a background color;
determine a boundary around the representation of the object in the image, a portion of the image outside the boundary and associated with the background color;
crop the image to generate a first cropped image to include only a portion of the image contained within the boundary;
locate a representation of a face in the first cropped image;
crop the first cropped image to generate a second cropped image, from an upper edge of the first cropped image, to remove a specified portion of the representation of the face from the image;
determine one or more regions in a lower half of the second cropped image including pixels with values corresponding to skin tones;
analyze rows of the pixels to locate features indicative of skin locations, and locate skin regions from the one or more regions of sufficient width associated with background color pixels outside the skin regions in a corresponding row of pixels;
determine a subset of the pixels in the skin regions, with values corresponding to skin tones, that correspond to a representation of one or more legs in the image;
crop the second cropped image to generate a third cropped image, from a lower edge of the second cropped image, to include only a specified portion of the representation of one or more legs in the third cropped image; and
provide the third cropped image for display.

2. The system of claim 1 , wherein the instructions when executed further cause the system to:
add columns of pixels of background color to the image in order to cause the image to have a specified aspect ratio.

3. The system of claim 1 , wherein the instructions when executed further cause the system to:
determine a representative color of the representation of the apparel item in the image; and
generate a new background color for the image based at least in part upon the background color and the representative color.

4. A computer-implemented method, comprising:
determining an object region, of a received image, including a representation of an object and a representation of an apparel item, the received image including a background associated with a first background color;
determining a location of a representation of a head in the received image;
determining a location of a representation of one or more legs in the received image, the representation of the one or more legs corresponding to pixels having skin tone color values;
using a center-weighted function to assign a relative weighting to pixels in the received image;
determining a number of pixels having a respective color, each pixel counting by an amount that corresponds to the relative weighting;
selecting the respective color with a highest number of pixels as a representative color of the representation of the apparel item in the received image;
generating a second image including the representation of the object; and
generating a second background color for the second image based at least in part on the first background color and the representative color, the second image excluding, relative to the received image: pixels of the second background color outside the object region, a first specified portion of the representation of the head, and a second specified portion of the representation of the one or more legs in the second image.

5. The computer-implemented method of claim 4 , further comprising:
generating a third image including additional pixels of background color in order to obtain a desired aspect ratio to the third image.

6. The computer-implemented method of claim 4 , further comprising:
scanning in, from a set of at least one of edges or corners of the received image, until one or more edges or contours of the representation of the object are located, in order to determine at least one of a foreground region or the object region.

7. The computer-implemented method of claim 4 , further comprising:
using at least one of a face detection algorithm or a head detection algorithm to detect the representation of the head in the received image; and
determine the first specified portion of the representation of the head to exclude based at least in part upon a confidence score generated by the at least one of the face detection algorithm or the head detection algorithm.

8. The computer-implemented method of claim 4 , further comprising:
generating a quantized color image having a lower color depth than the received image;
analyzing the quantized color image to locate regions of the pixels having the skin tone color values; and
generating a binary skin mask indicating the regions having the skin tone color values and regions not having skin-tone color values.

9. The computer-implemented method of claim 8 , further comprising:
scanning, from at least a middle row of the binary skin mask, rows of pixels to attempt to locate features corresponding to skin feature locations;
comparing the features to corresponding positions in a grayscale version of the received image to determine whether pixels of non-background color are located in corresponding rows of the received image; and
determining the location of the representation of the one or more legs at least in part by locating rows having skin feature regions of sufficient width, with only background color pixels outside the skin feature regions in respective rows of pixels.

10. The computer-implemented method of claim 4 , further comprising:
processing, with a classifier algorithm, a set of feature values for regions having skin tone color to determine the location of the representation of the one or more legs.

11. The computer-implemented method of claim 4 , further comprising:
reducing a color depth of the received image before determining the representative color.

12. The computer-implemented method of claim 4 , further comprising:
applying the new background color by at least one of shading a background region of the second image or shading an entirety of the second image with a minimum amount of transparency.

13. A non-transitory computer readable storage medium storing instructions that, when executed by at least one processor of a computing system, causes the computing system to:
determine an object region, of a received image, including a representation of an object and a representation of an apparel item, the received image including a background associated with a first background color;
determine a location of a representation of a head in the received image;
determine a location of a representation of one or more legs in the received image, the representation of the one or more legs corresponding to pixels having skin tone color values;
use a center-weighted function to assign a relative weighting to pixels in the received image;
determine a number of pixels having a respective color, each pixel counting by an amount that corresponds to the relative weighting;
select the respective color with a highest number of pixels as a representative color of the representation of the apparel item in the received image;
generate a second image including the representation of the object; and
generate a second background color for the second image based at least in part on the first background color and the representative color, the second image excluding, relative to the received image: pixels of

Assignees

• A9 Com Inc

Inventors

• Dhua Arnab Sanat Kumar

Classifications

• G06T11/10 Primary

  Texturing; Colouring; Generation of textures or colours (retouching, inpainting or scratch removal G06T5/77) · CPC title

• G06T2207/10024

  Color image · CPC title

• G06T11/60 Primary

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

• G06T11/001

  Physics · mapped topic

• G06T2210/36

  Level of detail · CPC title