1. Choose a texture
Use a PNG or JPG with even lighting. Painted shadows and highlights become grayscale variation too, so a softly lit texture usually gives a more useful starting point.
Roughness Map Generator
Turn a texture into an editable grayscale roughness map. Set the smooth-to-rough range, inspect the material response, and download an 8-bit PNG. Your image stays in your browser.
Up to 5 MB · 4096 px per side
See Pro file limitsBlack = smooth and sharper reflections. White = rough and broader reflections.
Source texture
Choose a texture to see its roughness map here.
Roughness map
This is a brightness-based starting point. A single photo cannot reveal a material’s true physical roughness; adjust the map under your scene’s lighting. Transparent pixels are exported as opaque grayscale.
How to make a roughness map from an image
2. Set the roughness range
Use Minimum to keep the smoothest parts from becoming mirror-like. Use Maximum to control the roughest parts. Contrast increases or reduces the variation within that range; Invert swaps the source’s dark and light regions.
3. Inspect and export
Open the 3D preview to inspect how highlights spread across the material. Download the PNG at your source dimensions, then test it under your own scene lighting.
What does a roughness map control?
Roughness controls the spread of reflected light, rather than surface height. Dark values produce smoother surfaces with tighter reflections; bright values produce rougher surfaces with broader reflections. The map does not add geometry, determine metalness, or replace a normal map.
Use the map in your material
In Blender, load the PNG as Non-Color data and connect its Color output to Principled BSDF → Roughness. In Unreal Engine, disable sRGB on the texture and connect it to the material’s Roughness input. These are material values, not display colors. For a workflow expecting smoothness or glossiness, use the inverse of roughness; do not connect the roughness PNG directly to a smoothness input.
Roughness, specular and normal maps
Use roughness to control the width of reflections, specular to control reflection intensity in a workflow that accepts a specular map, and normal maps to change apparent surface direction. They describe different properties. Inverting a specular map is not a reliable way to recover true roughness.
Roughness map questions
Can I generate an accurate roughness map from a photo?
You can generate a useful starting map from image brightness, but color, shadows and lighting are mixed in a photo. Review and adjust the result for the material instead of treating it as a physical measurement.
Is the roughness map black and white or colored?
The output is an opaque grayscale, 8-bit PNG: black represents smooth areas and white represents rough areas. It keeps the accepted source image dimensions.
Are my images uploaded?
No. Image decoding, roughness generation and PNG export run in your browser. The tool does not upload your source image.
What file sizes are supported?
Free supports PNG and JPG files up to 5 MB each; Pro supports up to 50 MB each. Both have a maximum of 4096 pixels per side. PNG export is available for free.