|Number of watchers on Github||50|
|Number of open issues||0|
|Average time to close an issue||about 1 hour|
|Open pull requests||0+|
|Closed pull requests||0+|
|Last commit||almost 5 years ago|
|Repo Created||almost 5 years ago|
|Repo Last Updated||over 2 years ago|
|Organization / Author||jcjohnson|
|Do you use texture-synthesis? Leave a review!|
|View texture-synthesis activity|
|View on github|
|Fresh, new opensource launches 🚀🚀🚀|
Software engineers: It's time to get promoted. Starting NOW! Subscribe to my mailing list and I will equip you with tools, tips and actionable advice to grow in your career.
This is a Torch implementation of a texture-synthesis algorithm very similar to .
Given an input texture patch, the algorithm generates a larger version of the same texture. Here's an example:
You can see more examples in the example gallery.
Texture synthesis is implemented in the script
synthesis.lua. The following command line options are available:
-source: Path to the source image.
-output_file: Path where the output should be written.
-height: Height of the output file, in pixels.
-width: Width of the output file, in pixels.
-k: Kernel size; must be an odd integer.
-gpu: Which GPU to use. Setting
gpu >= 0will run in GPU mode, and setting
gpu < 0will run in CPU-only mode.
You can see examples using different source images and kernel sizes in the example gallery.
 Efros, Alexei, and Thomas K. Leung.
Texture synthesis by non-parametric sampling. ICCV 1999.