Dear Geoscientists,
I am happy to announce ImageQuilting.jl, a geostatistics/computer vision package I’ve been working on since Julia v0.3. I’ve tested it extensively, but I would like to get your feedback if possible.
This package implements an extension to the famous Efros-Freeman algorithm for texture synthesis and transfer in computer vision. Unlike the original algorithm developed for 2D images, our method can also handle 3D grids and pre-existing point-data very efficiently (the fastest in the literature). This is useful for generating alternative models of the subsurface of the Earth when data is available at a few locations (e.g. wells) or at low resolution (e.g. seismic). You can find out more in the Examples section of the documentation.
The corresponding paper is coming out in a few days, I appreciate if you can report issues or give suggestions of improvement before it is out.
Hoffimann, J., Scheidt, C., Barfod, A., Caers, J, 2017. Stochastic Simulation by Image Quilting of Deterministic Process-based Geological Models.
I would like to thank the community for the constant help in forums and on GitHub, it is being a pleasure to learn from you. Special thanks to @tim.holy, @yuyichao, @vchuravy, and @tkelman.
Thank you,
-Júlio