Example-Based Color Transformation Image and Video Using Basic Color Categories Youngha Chang Suguru Saito Masayuki Nakajima
Dec 20, 2015
Example-Based Color Transformation of Image and Video Using Basic Color Categories
Youngha Chang
Suguru Saito
Masayuki Nakajima
ABSTRACT
the most effective method to improve the mood of an image
tradeoff between the quality and the manual operation
To achieve a more detailed and natural result with less labor, we suggest a method that performs an example-based color stylization of images using perceptual color categories.
OUTLINEOUTLINE
INTRODUCTION RELATED WORK APPROACH EXTENDING TO VIDEOS RESULTS
INTRODUCTION “color is more beautiful than shape”
one of the most important features in photo-editing or video postproduction tools
automatic manipulation & interactive tools
implement a method that transfers colors of an image with less user interaction
We consider that color stylization methods can be categorized into three categories:
1) shift all the pixel values in the same way
2) edit a specific image region with specific color values
3) adjust the color value of each color domain separately such that the adjustments do not cause any sense of incongruity
RELATED WORK Interactive tools can be categorized into three groups.
1) The first group edit an entire image in the same way. 2) The second group divide colors into six domains, R, Y, G, C, B,
and M, and control them independently. 3) The last group convert colors of user-defined image regions.
Example-based color transformation methods have recently been suggested.
We added a constraint during the color transformation process such that no pixel changes its original perceptual category after the color transformation.
basic color terms BCTs the achromatic ones : black, white, gray
the chromatic ones : red, green, yellow, blue, brown, purple, orange, and pink
RELATED WORK In this paper, we propose a technique that make the
algorithm more robust and to stylize the colors of video frame sequences.(1) Making the method applicable to images taken under a
variety of light conditions; (2) Speeding up the color-naming step;(3) Improving the mapping between source and reference
colors when there is a disparity in the size of the chromatic categories;
(4) Separate handling of achromatic categories from chromatic categories;(5) Extending the algorithm along the temporal axis to allow
video processing.
APPROACH
APPROACH
A. Handling the Illuminant Color in Images
A fundamental limitation of our previous algorithm is that it required target images to be taken under the D65.
We apply the color-by-correlation method to estimate the illuminant color.
APPROACH
B. Color Naming Method We denote the categorization as the “color naming” of a pixel.
The color naming method consists of two steps:
initial color naming and fuzzy color naming. Initial color naming, for example, if the color of a pixel is categorized
in the first BCC, then the vector becomes
In the initial color naming process, each pixel completely belongs to one of the 11 BCCs.
Fuzzy color naming is done to avoid pseudo-contours.
APPROACH
B. Color Naming Method
APPROACH
C. Computing Corresponding Color Values in the Chromatic Categories
APPROACH
D. Computing Corresponding Color Values In the Achromatic Categories
E. Transferring Colors
EXTENDING TO VIDEOS
A. Color Naming Method
EXTENDING TO VIDEOS
B. Computing Corresponding Color Values in the Chromatic Categories
EXTENDING TO VIDEOS
C. Computing Corresponding Color Values in the Achromatic Categories
D. Transferring Colors
RESULTS
RESULTS
RESULTS
RESULTS
RESULTS
Thank you for your listening!
2008.03.26