GLB & modeling
How to Convert a Photo to STL
By SnapTo3D · Published
Converting a photo to STL requires creating geometry from the image first. Choose a relief or lithophane for a picture-based surface, or a modeling or reconstruction workflow for a full object. Then prepare and export the geometry, confirm its size and inspect it in the intended slicer.
Choose the physical result first
A portrait plaque, a backlit picture panel and a freestanding character are three different outcomes. An image-to-STL tool may be designed for only one of them. Check examples of the geometry, not only the input and export extensions.
If you want a freestanding person, a rectangular relief with raised brightness values will not satisfy that requirement. Likewise, a full character-model workflow may be unnecessary when the goal is a flat decorative panel.
Route one: relief from image values
A relief uses image information to control the height of a surface. Blender’s Displace modifier, for example, can move mesh vertices according to texture intensity. This provides a technical building block, not automatic recovery of the pictured object’s anatomy.
Brightness is not measured depth. A dark shirt can be pushed down while a bright background rises unless the image is prepared for the chosen mapping. Inspect the height interpretation and deliberately design the backing and edges.
Route two: a lithophane
A lithophane uses differences in material thickness to form an image when illuminated from behind. It is a physical image panel, not a complete reconstruction of the subject.
Follow the chosen generator and printing material’s guidance for thickness and orientation, then validate the result with the intended light. This guide supplies no universal thickness or printer settings and makes no claim of a successful test print.
Route three: a full model
For a freestanding object, use manual modeling, single-image generation or an appropriate multi-photo reconstruction workflow. A single photo does not reveal the rear or hidden surfaces, so a generative route must infer those regions.
SnapTo3D’s photo-based character route produces a stylized interpretation rather than a measured scan. If using an eligible GLB download, convert it externally only after confirming the receiving application can import the asset and its compression features.
Prepare the geometry before exporting
Inspect the result for missing surfaces, unintended fragments and features too thin for the selected process. Set a physical target size and consider whether the object needs a base or other design changes. Exporting STL does not make those decisions for you.
For a relief, check that the intended volume has a backing and closed boundary. For a full character, inspect the silhouette from every side and review disconnected details. Keep the editable source before applying repairs.
Validate the handoff in the slicer
Import the STL into the intended slicer and verify dimensions instead of assuming the exporter and slicer interpreted units identically. Inspect the layer preview for missing details, gaps and unexpected islands.
Retain the source photo, original model and export separately. If texture color mattered in the digital preview, do not assume it survives as standard STL appearance. Confirm the printer or service’s actual requirements before ordering or printing.
Frequently asked questions
Can I convert a JPG to STL by changing its filename?
No. A JPG contains pixels; an STL needs surface geometry. A modeling or image-to-geometry step must happen first.
Will a relief recover the person’s real facial depth?
Not automatically. A brightness-to-height mapping is an interpretation of image values, not a measurement of facial geometry.
Is photo-to-STL the same as photo-to-figurine?
No. The result may be a relief or lithophane rather than a freestanding character. Confirm the intended physical object.