A multiresolution spline with application to image mosaics
ACM Transactions on Graphics (TOG)
Graphcut textures: image and video synthesis using graph cuts
ACM SIGGRAPH 2003 Papers
ACM SIGGRAPH 2003 Papers
Interactive digital photomontage
ACM SIGGRAPH 2004 Papers
ACM SIGGRAPH 2004 Papers
ACM SIGGRAPH 2006 Papers
A Closed Form Solution to Natural Image Matting
CVPR '06 Proceedings of the 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Volume 1
Random Walks for Image Segmentation
IEEE Transactions on Pattern Analysis and Machine Intelligence
Content-aware copying and pasting in images
The Visual Computer: International Journal of Computer Graphics
Seamless video composition using optimized mean-value cloning
The Visual Computer: International Journal of Computer Graphics
Environment-Sensitive cloning in images
The Visual Computer: International Journal of Computer Graphics - CGI'2011 Conference
Hi-index | 0.03 |
Seamless image composition is a widely-used technique for digital image re-creation. Poisson image editing is an effective approach for seamless editing using guided interpolation. However, when applied to image composition with quite different tone in source and target images, like distinct illumination or color distribution, its performance often degrades rapidly owing to deficient user-drawn boundary and immoderate color transition. In this paper, we propose a new solution based on Poisson image editing to keep the coherent tone in the composition result. We firstly employ matting technique to help user effortlessly locate the object of interest. Meanwhile, we build transition zones to coordinate the influence from source and target images with the matte from matting techniques. Then we construct a mixed guided vector based on transition zones to implement the tone adjustment. An effective method is also proposed to find the optimal mixed guided vector. At last, we utilize this optimal vector into Poisson solution. Our method can avoid the drawbacks of Poisson image editing between scenes with distinct tone and produce enjoyable composition results. Experiment results demonstrate the effectiveness of our seamless image composition method.