Image cropping

Patent metadata

Field	Value
Publication number	US-2016189343-A1
Application number	US-201414588213-A
Country	US
Kind code	A1
Filing date	Dec 31, 2014
Priority date	Dec 31, 2014
Publication date	Jun 30, 2016
Grant date	—

Abstract

Briefly, embodiments disclosed herein relate to image cropping, such as for digital images, for example.

First claim

1 . A method of executing computer instructions on at least one computing device without further human interaction in which the at least one computing device includes at least one processor and at least one memory, comprising:
fetching computer instructions from the at least one memory of the at least one computing device for execution on the at least one processor of the at least one computing device; executing the fetched computer instructions on the at least one processor of the at least one computing device; and storing in the at least one memory of the at least one computing device any results of having executed the fetched computer instructions on the at least one processor of the at least one computing device; wherein the computer instructions to be executed comprise instructions for determining a cropping window for a region of an image; wherein the executing the fetched instructions further comprises: generating a composite saliency map for the region of the image; and determining the cropping window for the region of the image based, at least in part, on the composite saliency map.

2 . The method of claim 1 , wherein the image comprises a digital image.

3 . The method of claim 1 , wherein the generating the composite saliency map comprises generating a plurality of pre-composite saliency maps.

4 . The method of claim 3 , wherein the plurality of pre-composite saliency maps comprise a saliency map to model human eye-attention in images, a face saliency map, a position saliency map, or any combination thereof.

5 . The method of claim 1 , further comprising determining the region of the image, wherein the region comprises a contiguous portion of the image.

6 . The method of claim 1 , further comprising determining the region of the image, wherein the region comprises non-contiguous portions of the image.

7 . The method of claim 1 , further comprising:
detecting one or more separating boundaries in the image; and determining the region based, at least in part, on the one or more detected separating boundaries.

8 . The method of claim 7 , wherein the determining the region comprises identifying a candidate region having an aspect ratio and having a width and/or height related to the aspect ratio, and further comprises determining whether the width and/or height of the candidate region exceeds a threshold.

9 . The method of claim 8 , wherein the threshold is specified at least in part based on a width and/or height of the image.

10 . The method of claim 1 , wherein the image comprises an original image.

11 . The method of claim 8 ,
wherein the separating boundaries indicate one or more borders between two or more regions, and wherein the determining the region further comprises selecting the region from the two or more regions at least in part in response to a determination that the width and/or height of the candidate region does not exceed the threshold.

12 . The method of claim 8 , further comprising:
wherein the determining the region comprises selecting the candidate region as the region at least in part in response to a determination that the width and/or height of the candidate region exceeds the threshold.

13 . The method of claim 12 , wherein the threshold comprises a user specified threshold, a programmable threshold, or an adaptive threshold, or any combination thereof.

14 . An apparatus, comprising:
at least one computing device; the at least one computing device to include at least one processor and at least one memory; the at least one computing device to execute computer instructions on the at least one processor without further human intervention; the computer instructions to be executed to have been fetched from the at least one memory for execution on the at least one processor, and the at least one computing device to store in the at least one memory of the at least one computing device any results to be generated from the execution on the at least one processor of the to be executed computer instructions; the computer instructions to be executed to comprise instructions to determine a cropping window for a region of an image; wherein the instructions to be executed to: generate a composite saliency map for the region of the image; and determine the cropping window for the region of the image based, at least in part, on the composite saliency map.

15 . The apparatus of claim 14 , wherein the at least one processor to compute a plurality of saliency maps to compute the composite saliency map.

16 . The apparatus of claim 15 , wherein the plurality of saliency maps to comprise a saliency map to model human eye-attention in images, a face saliency map, a position saliency map, or any combination thereof.

17 . The apparatus of claim 14 , wherein the at least one processor further to determine the region of the image, wherein the region to comprise a contiguous portion of the image.

18 . The apparatus of claim 14 , the processor further to determine the region of the image, wherein the region comprises a non-contiguous portion of the image.

19 . The apparatus of claim 14 , wherein the at least one processor to detect one or more separating boundaries in the image and to determine the region of the image based, at least in part, on the one or more separating boundaries to be detected.

20 . The apparatus of claim 14 , wherein the at least one processor to identify a candidate region to have an aspect ratio and to have a width and/or height to be related to the aspect ratio, and wherein the at least one processor further to determine whether the width and/or height of the candidate region exceeds a threshold.

21 . An apparatus, comprising:
means for executing computer instructions on at least one computing device without further human interaction in which the at least one computing device includes at least one processor and at least one memory, comprising: means for fetching computer instructions from the at least one memory of the at least one computing device for execution on the at least one processor of the at least one computing device; means for executing the fetched computer instructions on the at least one processor of the at least one computing device; and means for storing in the at least one memory of the at least one computing device any results of having executed the fetched computer instructions on the at least one processor of the at least one computing device; wherein the computer instructions comprise instructions for determining a cropping window for a region of a image; wherein the means for executing the fetched instructions further comprises: means for generating a composite saliency map for the region of the image; and means for determining the cropping window for the region of the image based, at least in part, on the composite saliency map.

Assignees

• Yahoo Inc

Inventors

• Chen Daozheng
• Lee Sally
• Webb Brian
• Rabbat Ralph
• Khodaei Ali
• Krakow Paul
• Todd Dave
• Giordano Samantha
• Chern Max

Classifications

• G06V10/803

  of input or preprocessed data · CPC title

• G06V10/40

  Extraction of image or video features · CPC title

• G06F18/251

  of input or preprocessed data · CPC title

• G06V10/25

  Determination of region of interest [ROI] or a volume of interest [VOI] · CPC title

• G06K2009/4666

  Physics · mapped topic