IEEE Computer Graphics and Applications
Computer Aided Geometric Design
Colorization using optimization
ACM SIGGRAPH 2004 Papers
Local Color Transfer via Probabilistic Segmentation by Expectation-Maximization
CVPR '05 Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) - Volume 1 - Volume 01
Mean value coordinates for closed triangular meshes
ACM SIGGRAPH 2005 Papers
Interactive local adjustment of tonal values
ACM SIGGRAPH 2006 Papers
Visual attention detection in video sequences using spatiotemporal cues
MULTIMEDIA '06 Proceedings of the 14th annual ACM international conference on Multimedia
Automated colour grading using colour distribution transfer
Computer Vision and Image Understanding
ACM SIGGRAPH 2007 papers
PG '07 Proceedings of the 15th Pacific Conference on Computer Graphics and Applications
AppProp: all-pairs appearance-space edit propagation
ACM SIGGRAPH 2008 papers
Coordinates for instant image cloning
ACM SIGGRAPH 2009 papers
Efficient affinity-based edit propagation using K-D tree
ACM SIGGRAPH Asia 2009 papers
Progressive histogram reshaping for creative color transfer and tone reproduction
NPAR '10 Proceedings of the 8th International Symposium on Non-Photorealistic Animation and Rendering
Diffusion maps for edge-aware image editing
ACM SIGGRAPH Asia 2010 papers
Fast local color transfer via dominant colors mapping
ACM SIGGRAPH ASIA 2010 Sketches
Soft Color Segmentation and Its Applications
IEEE Transactions on Pattern Analysis and Machine Intelligence
| Hi-index | 0.00 |
Interactive color transfer is an important image editing technique, which changes the color pattern of a given source image following users' intention and is more flexible than fully automatic methods in control of color modification. This paper presents a new interactive transfer approach which consists of two stages. Firstly, users specify the color change style by drawing some strokes in the source image and their reference strokes in one or multiple reference images. The source strokes should respectively obey the color style of their reference strokes after color transferring. This is actually a color transfer problem too, which is then casted as a quadratic optimization forcing the color statistic attributes of the source strokes to follow those of their reference strokes while preserving their gradient map unchanged. The source strokes are separately treated and therefore they can bear different color styles after processing. The solution of the optimization is very efficient due to the fact that the specified strokes with the same color style usually contains a small number of pixels. Secondly, the new colors of pixels in source strokes are propagated to the whole image automatically. Instead of solving an optimization equation, we renew the color attributes of each pixel in the source image using a scattered interpolation function which is expressed as a weighted combination of the new colors of pixels in those specified strokes, where the weight is inverse to the multiplication of the appearance distance and position distance from the current pixel to the pixel in the source strokes. Experimental results show that the proposed approach can not only preserve the texture detail and reflect user's intention well but also achieve realtime performance even for large images.