pOp: Parameter Optimization of Differentiable Vector Patterns

Teaser image — pOp finds the parameters of procedural vector graphics patterns that best match target images. We support patterns comprised of standard vector graphics elements, e.g. circles, rectangles, lines, and quadratic Bèzier curves, where the translation, rotation and scale of the elements is defined by an arbitrary procedural program. Here we show several examples from different generators. We tested our algorithm with synthetic input generated from a pattern instance, that lets us measure the goodness of fit, here reported as mean squared error (MSE) of the procedural parameters. We also include examples fitted from hand-drawn input, mimicking a possible design application.

Paper

Abstract

Procedural materials are extensively used in computer graphics, since they provide editable, resolution-independent representation of textures. However, tuning the parameters of procedural generators to achieve a desired result remains time-consuming for users. Recently, inverse procedural material algorithms have been developed, exploiting differentiable rendering methods to find the parameters of a procedural model that match a target image. These approaches focus on raster textures. We propose pOp, a practical method for estimating the parameters of vector patterns, that are formed by collections of vector shapes arranged by an arbitrary procedural program. In our approach, patterns are defined as arbitrary programs, that control the translation, rotation and scale or vector graphics elements. We support elements typical of vector graphics, namely points, lines, circle, rounded rectangles, and quadratic Bèzier drawings, in multiple colors. We optimize the program parameters by automatically differentiating the signed distance field of the drawing, which we found to be significantly more reliable than using differentiable rendering of the final image. We demonstrate our method on a variety of cases, representing the variations found in structured vector patterns.

Citation

@article{riso22pop, author = {Riso, Marzia and Sforza, Davide and Pellacini, Fabio}, title = {pOp: Parameter Optimization of Differentiable Vector Patterns}, journal = {Computer Graphics Forum}, volume = {41}, number = {4}, pages = {161-168}, keywords = {CCS Concepts, • Computing methodologies → Texturing}, doi = {https://doi.org/10.1111/cgf.14595}, url = {https://onlinelibrary.wiley.com/doi/abs/10.1111/cgf.14595}, eprint = {https://onlinelibrary.wiley.com/doi/pdf/10.1111/cgf.14595}, year = {2022} }