Xingchang Huang1, Pooran Memari2, Hans-Peter Seidel1, Gurprit Singh1
1Max Planck Institute for Informatics, Saarbrücken, 2 CNRS, LIX, Ecole Polytechnique, Paris, France
EGSR 2022 / Computer Graphics Forum, Volume 41 issue 6, July 2022
Our method takes simply a point set (with positions, classes, attributes) as input and applies continuous Gabor transform to extract
features. We then use these Gabor features to perform pattern expansion for a large canvas. We show synthesis results of a 2-class point pattern in (a), and a
4-class point pattern with depth and scale as attributes in (b).
Abstract
Point pattern synthesis requires capturing both local and non-local correlations from a given exemplar. Recent works employ
deep hierarchical representations from VGG-19 convolutional network to capture the features for both point-pattern and
texture synthesis. In this work, we develop a simplified optimization pipeline that uses more traditional Gabor transform-based
features. These features when convolved with simple random filters gives highly expressive feature maps. The resulting framework
requires significantly less feature maps compared to VGG-19-based methods, better captures both the
local and non-local structures, does not require any specific data set training and can easily extend to handle multi-class and
multi-attribute point patterns, e.g., disk and other element distributions. To validate our pipeline, we perform qualitative and
quantitative analysis on a large variety of point patterns to demonstrate the effectiveness of our approach. Finally, to better
understand the impact of random filters, we include a spectral analysis using filters with different frequency bandwidths.
