Proceedings of the 27th annual conference on Computer graphics and interactive techniques
Reconstruction and representation of 3D objects with radial basis functions
Proceedings of the 28th annual conference on Computer graphics and interactive techniques
A Practical Image Retouching Method
CW '02 Proceedings of the First International Symposium on Cyber Worlds (CW'02)
RBF-based image restoration utilising auxiliary points
Proceedings of the 2009 Computer Graphics International Conference
| Hi-index | 0.00 |
Radial basis functions (RBF) based methods are widely used for interpolation of scattered data due to their abilities, so the principles could be also effectively used on reconstruction of damaged images. This paper deals with study of influence of some essential parameters to the image reconstruction using RBF. We focus mainly on parameters such as size of the involved neighbourhood for the reconstruction, various extending elements of the linear system and finally the choice of used radial basis function itself. We also present a new algorithm for reconstructing damaged images. There we focus mainly on getting the best possible visual results of the reconstruction. Our new approach will use RBF principles to calculate damaged pixels, so some time consuming calculations could occur depending on damage character and amount. We test this new algorithm on non-trivial real data damaged by defects of various characters and compare it with another RBF based algorithm.