Title: | Blending textured images using a non-parametric multiscale MRF method |
Authors: | Tiddeman, Bernard |
Citation: | Journal of WSCG. 2004, vol. 12, no. 1-3, p. 459-465. |
Issue Date: | 2004 |
Publisher: | UNION Agency |
Document type: | článek article |
URI: | http://wscg.zcu.cz/wscg2004/Papers_2004_Full/K29.pdf http://hdl.handle.net/11025/1745 |
ISSN: | 1213-6972 |
Keywords: | obličejové prototypování;algoritmus Markovova náhodného pole;texturní syntéza |
Keywords in different language: | facial prototyping;Markov random field algorithm;texture synthesis |
Abstract: | In this paper we describe a new method for improving the representation of textures in blends of multiple images based on a Markov Random Field (MRF) algorithm. We show that direct application of an MRF texture synthesis algorithm across a set of images is unable to capture both the "averageness" of the global image appearance as well as specific textural components. To overcome this problem we vary the width of the Parzen window (used to smooth the conditional probability distribution of the pixel's intensity) as a function of scale, thus making lower pyramid resolutions closer to the Gaussian mean, while maintaining the high resolution textures. We also show that approximating the maxima of the conditional probability distributions with a weighted-average produces very similar results with a significant increase in speed. |
Rights: | © UNION Agency |
Appears in Collections: | Volume 12, number 1-3 (2004) |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
K29.pdf | 711,64 kB | Adobe PDF | View/Open | |
K29-1.zip | 574,92 kB | Unknown | View/Open |
Please use this identifier to cite or link to this item:
http://hdl.handle.net/11025/1745
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.