Method and apparatus for generating image filter

Patent metadata

Field	Value
Publication number	US-9959484-B2
Application number	US-201615015785-A
Country	US
Kind code	B2
Filing date	Feb 4, 2016
Priority date	Feb 11, 2015
Publication date	May 1, 2018
Grant date	May 1, 2018

Abstract

The present disclosure discloses a method and an apparatus for generating an image filter. The method includes receiving a first image that is selected with a specific imaging effect desired for the image filter, generating a group of images corresponding to the first image, wherein the group of images comprises at least a second image of a common structure characteristic with the first image and without the specific imaging effect, calculating a mapping relationship for transforming the group of images to the first image and storing the mapping relationship to be used as the image filter to generate the specific imaging effect on other images.

First claim

What is claimed is:

1. A method for generating an image filter, wherein the method is applied in an electronic device, and the method comprises:
receiving, by the electronic device, a first image that is selected with a specific imaging effect desired for the image filter;
generating, by the electronic device, a group of images corresponding to the first image, wherein the group of images comprises at least a second image of a common structure characteristic with the first image and the second image does not have the specific imaging effect and is selected from an image library;
calculating, by the electronic device, a mapping relationship for transforming the group of images to the first image; and
storing, in the electronic device, the mapping relationship to be used as the image filter to generate the specific imaging effect on other images,
wherein generating, by the electronic device, the group of images corresponding to the first image comprises:
calculating, by the electronic device, a first feature vector of the first image;
obtaining, by the electronic device, a second feature vector of each sample image in the image library, the sample images in the image library being images without the specific imaging effect;
comparing, by the electronic device, the first feature vector with each second feature vector to determine, from the sample images, at least the second image of the common structure characteristic with the first image; and
forming, by the electronic device, the group of images including at least the second image.

2. The method of claim 1 , wherein calculating, by the electronic device, the mapping relationship for transforming the group of images to the first image, comprises:
calculating, by the electronic device, a first color mean value and a first color variance of pixels within each image in the group of images;
calculating, by the electronic device, a second color mean value and a second color variance of pixels within the first image; and
generating, by the electronic device, the mapping relationship based on the first color mean value, the first color variance, the second color mean value, and the second color variance.

3. The method of claim 2 , wherein the mapping relationship is generated by the electronic device as:

{

L
=

(

vG_l
/
va_l

)

×

(

l
-
mA_l

)

+
mG_l

A
=

(

vG_a
/
vA_a

)

×

(

l
-
mA_a

)

+
mG_a

B
=

(

vG_b
/
vA_b

)

×

(

l
-
mA_b

)

+
mG_b

where (l, a, b) denotes to a pixel value of a pixel before transforming, and (L, A, B) denotes to a pixel value of the pixel after transforming, (mG_l, mG_a, mG_b) denotes the first color mean value, (vG_l, vG_a, vG_b) denotes to the first color variance, (mA_l, mA_a, mA_b) denotes the second color mean value, (vA_l, vA_a, vA_b) denotes to the second color variance.

4. The method of claim 1 , wherein comparing, by the electronic device, the first feature vector with each second feature vector to determine, from the sample images, at least the second image of the common structure characteristic to the first image further comprises:
calculating, by the electronic device, a distance between the first feature vector of the first image and a second feature vector of a sample image; and
determining, by the electronic device, the sample image of the common structure characteristic to the first image when the distance is smaller than a threshold.

5. The method of claim 1 , wherein calculating, by the electronic device, the first feature vector of the first image further comprises:
sub-sampling, by the electronic device, the first image to generate a pyramid representation of the first image at multiple scales; and
calculating, by the electronic device, the first feature vector based on the pyramid representation.

6. An apparatus for generating an image filter, wherein the apparatus is included in an electronic device, and the apparatus comprises:
a processor; and
a memory for storing instructions executable by the processor;
wherein the processor is configured to:
receive a first image that is selected with a specific imaging effect desired for the image filter;
generate a group of images corresponding to the first image, wherein the group of images comprises at least a second image of a common structure characteristic with the first image and the second image does not have the specific imaging effect and is selected from an image library;
calculate a mapping relationship for transforming the group of images to the first image; and
store the mapping relationship to be used as the image filter in the memory,

wherein the processor is further configured to calculate a first feature vector of the first image, obtain a second feature vector of each sample image in the image library, the sample image being an im

Assignees

• Xiaomi Inc

Inventors

• Wang Baichao
• Chen Zhijun
• Hou Wendi

Classifications

• H04N1/60 Primary

  Colour correction or control {(H04N1/54 takes precedence)} · CPC title

• G06T5/50

  using two or more images, e.g. averaging or subtraction · CPC title

• G06K9/481

  Physics · mapped topic

• G06T2207/10024

  Color image · CPC title

• G06T2207/20024

  Filtering details · CPC title