TY - JOUR
T1 - Parametric texture model based on joint statistics of complex wavelet coefficients
AU - Portilla, Javier
AU - Simoncelli, Eero P.
N1 - Funding Information:
JP is partially supported by a fellowship from the Pro-grama Nacional de Formacion de Personal Investigador (Spanish Government). EPS is supported by an Alfred P. Sloan Research Fellowship, NSF CAREER grant MIP-9796040, and the Sloan Center for Theoretical Neurobiology at NYU. The authors wish to thank Ted Adelson and Jonathan Victor for interesting discussions, and the reviewers for helpful comments.
PY - 2000/10
Y1 - 2000/10
N2 - We present a universal statistical model for texture images in the context of an overcomplete complex wavelet transform. The model is parameterized by a set of statistics computed on pairs of coefficients corresponding to basis functions at adjacent spatial locations, orientations, and scales. We develop an efficient algorithm for synthesizing random images subject to these constraints, by iteratively projecting onto the set of images satisfying each constraint, and we use this to test the perceptual validity of the model. In particular, we demonstrate the necessity of subgroups of the parameter set by showing examples of texture synthesis that fail when those parameters are removed from the set. We also demonstrate the power of our model by successfully synthesizing examples drawn from a diverse collection of artificial and natural textures.
AB - We present a universal statistical model for texture images in the context of an overcomplete complex wavelet transform. The model is parameterized by a set of statistics computed on pairs of coefficients corresponding to basis functions at adjacent spatial locations, orientations, and scales. We develop an efficient algorithm for synthesizing random images subject to these constraints, by iteratively projecting onto the set of images satisfying each constraint, and we use this to test the perceptual validity of the model. In particular, we demonstrate the necessity of subgroups of the parameter set by showing examples of texture synthesis that fail when those parameters are removed from the set. We also demonstrate the power of our model by successfully synthesizing examples drawn from a diverse collection of artificial and natural textures.
UR - http://www.scopus.com/inward/record.url?scp=0034291204&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0034291204&partnerID=8YFLogxK
U2 - 10.1023/A:1026553619983
DO - 10.1023/A:1026553619983
M3 - Article
AN - SCOPUS:0034291204
SN - 0920-5691
VL - 40
SP - 49
EP - 71
JO - International Journal of Computer Vision
JF - International Journal of Computer Vision
IS - 1
ER -