Synthesizing Structured Image Hybrids

Eric Risser 1 Charles Han 2 Rozenn Dahyot1 Eitan Grinspun2

1Trinity College Dublin 2Columbia University

Abstract
----------------------------------------------------------------------------------------------------------------------------------------------------------------

Example-based texture synthesis algorithms generate novel texture images from example data. A popular hierarchical pixel-based approach uses spatial jitter to introduce diversity, at the risk of breaking coarse structure beyond repair. We propose a multiscale descriptor that enables appearance-space jitter, which retains structure. This idea enables repurposing of existing texture synthesis implementations for a qualitatively different problem statement and class of inputs: generating hybrids of structured images.

Files
----------------------------------------------------------------------------------------------------------------------------------------------------------------

PAPER: [PDF, 19MB]

SLIDES: [PPT + Videos, 83MB]

VIDEO: [Descriptor Space, 66MB]

BibTeX
----------------------------------------------------------------------------------------------------------------------------------------------------------------

@article{RHDG10,

author = {Eric Risser and Charles Han and Rozenn Dahyot and Eitan Grinspun},
title = {Synthesizing Structured Image Hybrids},
journal = {ACM Transactions on Graphics (Proceedings of SIGGRAPH 2010)},
year = {2010},
volume = {29},
number = {4},
pages = {85:1--85:6}

}


----------------------------------------------------------------------------------------------------------------------------------------------------------------