In digital cinema, an image goes through many types of processes like scanning, mastering, and digital intermediate. Among them, the digital intermediate process plays a central role because it determines the final color of an image. It edits and changes the colors of the images. However, some color distortions such as color bleeding are generated when editing and changing local colors in an image. In this paper, local color improvement for digital intermediate is proposed based on color transfer. Our method is simple and efficient color improvement that does not requires neither precise image segmentation nor feature matching. To prevent color distortions, a modified color influence map is proposed with color categories. First, the source image is roughly segmented using a color category map, which groups similar colors in color space. Second, the color influence map is modified by assigning different weights to the lightness and chroma components. Lastly, the modified color influence map and color category map filtered with anisotropic diffusion are combined. Experimental results show that the proposed method produces less color distortion in the resulting image.
Ho-Gun HA
Kyungpook National University
Dae-Chul KIM
Kyungpook National University
Wang-Jun KYUNG
Kyungpook National University
Yeong-Ho HA
Kyungpook National University
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Ho-Gun HA, Dae-Chul KIM, Wang-Jun KYUNG, Yeong-Ho HA, "Color Improvement Based on Local Color Transfer in Digital Intermediate of Digital Cinema" in IEICE TRANSACTIONS on Information,
vol. E98-D, no. 1, pp. 173-179, January 2015, doi: 10.1587/transinf.2014EDP7179.
Abstract: In digital cinema, an image goes through many types of processes like scanning, mastering, and digital intermediate. Among them, the digital intermediate process plays a central role because it determines the final color of an image. It edits and changes the colors of the images. However, some color distortions such as color bleeding are generated when editing and changing local colors in an image. In this paper, local color improvement for digital intermediate is proposed based on color transfer. Our method is simple and efficient color improvement that does not requires neither precise image segmentation nor feature matching. To prevent color distortions, a modified color influence map is proposed with color categories. First, the source image is roughly segmented using a color category map, which groups similar colors in color space. Second, the color influence map is modified by assigning different weights to the lightness and chroma components. Lastly, the modified color influence map and color category map filtered with anisotropic diffusion are combined. Experimental results show that the proposed method produces less color distortion in the resulting image.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2014EDP7179/_p
Copy
@ARTICLE{e98-d_1_173,
author={Ho-Gun HA, Dae-Chul KIM, Wang-Jun KYUNG, Yeong-Ho HA, },
journal={IEICE TRANSACTIONS on Information},
title={Color Improvement Based on Local Color Transfer in Digital Intermediate of Digital Cinema},
year={2015},
volume={E98-D},
number={1},
pages={173-179},
abstract={In digital cinema, an image goes through many types of processes like scanning, mastering, and digital intermediate. Among them, the digital intermediate process plays a central role because it determines the final color of an image. It edits and changes the colors of the images. However, some color distortions such as color bleeding are generated when editing and changing local colors in an image. In this paper, local color improvement for digital intermediate is proposed based on color transfer. Our method is simple and efficient color improvement that does not requires neither precise image segmentation nor feature matching. To prevent color distortions, a modified color influence map is proposed with color categories. First, the source image is roughly segmented using a color category map, which groups similar colors in color space. Second, the color influence map is modified by assigning different weights to the lightness and chroma components. Lastly, the modified color influence map and color category map filtered with anisotropic diffusion are combined. Experimental results show that the proposed method produces less color distortion in the resulting image.},
keywords={},
doi={10.1587/transinf.2014EDP7179},
ISSN={1745-1361},
month={January},}
Copy
TY - JOUR
TI - Color Improvement Based on Local Color Transfer in Digital Intermediate of Digital Cinema
T2 - IEICE TRANSACTIONS on Information
SP - 173
EP - 179
AU - Ho-Gun HA
AU - Dae-Chul KIM
AU - Wang-Jun KYUNG
AU - Yeong-Ho HA
PY - 2015
DO - 10.1587/transinf.2014EDP7179
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E98-D
IS - 1
JA - IEICE TRANSACTIONS on Information
Y1 - January 2015
AB - In digital cinema, an image goes through many types of processes like scanning, mastering, and digital intermediate. Among them, the digital intermediate process plays a central role because it determines the final color of an image. It edits and changes the colors of the images. However, some color distortions such as color bleeding are generated when editing and changing local colors in an image. In this paper, local color improvement for digital intermediate is proposed based on color transfer. Our method is simple and efficient color improvement that does not requires neither precise image segmentation nor feature matching. To prevent color distortions, a modified color influence map is proposed with color categories. First, the source image is roughly segmented using a color category map, which groups similar colors in color space. Second, the color influence map is modified by assigning different weights to the lightness and chroma components. Lastly, the modified color influence map and color category map filtered with anisotropic diffusion are combined. Experimental results show that the proposed method produces less color distortion in the resulting image.
ER -