HugoTini/DeepBump

HugoTini/DeepBump has 1,425 GitHub stars and 64 forks. GitHot measured 0 stars in 24 hours. Last checked 2026-11-08 05:38 UTC. Last push 2025-03-17.

Normal & height maps generation from single pictures

1,425 stars 64 forks

Measured star growth

Net star changes over overlapping periods, not daily totals. Missing measurements are left blank. Stars measure attention, not software quality.

DeepBump is a machine-learning driven tool to generate normal & height maps from single pictures. See this blog post for an introduction.

Download DeepBump as a ZIP.
In Blender, go to Edit -> Preferences -> Add-ons -> Top right arrow -> Install from disk and select the downloaded ZIP file. Then enable the add-on.
In the add-on preference, click the 'Install dependencies' button (this requires an internet connection and might take a while). In case of error, try running Blender as administrator for this step.

By installing those dependencies, be aware of Microsoft conditions. This add-on uses available APIs to disable telemetry.

Check the blog post first video.

In the Shader Editor, in the side bar under the DeepBump tab:

Select a color image node in the shader editor, then click Generate Normal Map.

Select the generated normal map image node in the shader editor, then click Generate Height Map.

Select the generated normal map image node in the shader editor, then click Generate Curvature Map.

Select a color image node in the shader editor, then click Upscale.

For command-line usage, see cli.md.

This repo is under the GPL license. The training code is currently not available.