TY - JOUR
T1 - On neighbourhood matching for Texture-by-Numbers
AU - Sivaks, Eliyahu
AU - Lischinski, Dani
PY - 2011/3
Y1 - 2011/3
N2 - Texture-by-Numbers is an attractive texture synthesis framework, because it is able to cope with non-homogeneous texture exemplars, and provides the user with intuitive creative control over the outcome of the synthesis process. Like many other exemplar-based texture synthesis methods, its basic underlying mechanism is neighbourhood matching. In this paper we review a number of commonly used neighbourhood matching acceleration techniques, compare and analyse their performance in the specific context of Texture-by-Numbers (as opposed to ordinary unconstrained texture synthesis). Our study indicates that the standard approaches are not optimally suited for the Texture-by-Numbers framework, often producing visually inferior results compared to searching for the exact L2nearest neighbour. We then show that performing Texture-by-Number using the Texture Optimization framework in conjunction with an efficient FFT-based search is able to produce good results in reasonable running times and with a minimal memory overhead.
AB - Texture-by-Numbers is an attractive texture synthesis framework, because it is able to cope with non-homogeneous texture exemplars, and provides the user with intuitive creative control over the outcome of the synthesis process. Like many other exemplar-based texture synthesis methods, its basic underlying mechanism is neighbourhood matching. In this paper we review a number of commonly used neighbourhood matching acceleration techniques, compare and analyse their performance in the specific context of Texture-by-Numbers (as opposed to ordinary unconstrained texture synthesis). Our study indicates that the standard approaches are not optimally suited for the Texture-by-Numbers framework, often producing visually inferior results compared to searching for the exact L2nearest neighbour. We then show that performing Texture-by-Number using the Texture Optimization framework in conjunction with an efficient FFT-based search is able to produce good results in reasonable running times and with a minimal memory overhead.
KW - energy minimization
KW - example-based rendering
KW - image analogies
KW - texture optimization
KW - texture synthesis
KW - texture-by-numbers
UR - http://www.scopus.com/inward/record.url?scp=84861993308&partnerID=8YFLogxK
U2 - 10.1111/j.1467-8659.2010.01837.x
DO - 10.1111/j.1467-8659.2010.01837.x
M3 - ???researchoutput.researchoutputtypes.contributiontojournal.systematicreview???
AN - SCOPUS:84861993308
SN - 0167-7055
VL - 30
SP - 127
EP - 138
JO - Computer Graphics Forum
JF - Computer Graphics Forum
IS - 1
ER -