Photogrammetry, lithophane, or image to 3D? Three ways to get a printable model from a picture

Turning a photo into a printable 3D model can be accomplished through three distinct methods. The first approach involves lithophanes and reliefs, where a single photo is converted into a thin plate with varying thickness based on the brightness, creating a visual effect when viewed from an angle. This technique is ideal for capturing the essence of a photo or producing personalized gifts but does not result in a true three-dimensional object.

The second method is photogrammetry, which involves capturing multiple overlapping images of a physical object and using specialized software to reconstruct its geometry digitally. Photogrammetry is perfect for reproducing existing objects accurately, preserving their proportions, and generating functional parts. However, it may struggle with shiny, transparent surfaces and needs careful cleanup to achieve the best results.

The third method, image-to-3D conversion, is a more recent approach that leverages artificial intelligence to generate a complete 3D mesh from a single image or even a textual description. This technique is particularly useful when you have an idea or a sketch of an object and want to create a tangible representation. However, the output may not always be precise, especially for functional parts, and additional post-processing may be required to ensure accuracy.

To help users decide which method to use, a quick decision table is provided. It suggests lithophane generation for photos meant to be appreciated visually, photogrammetry for accurate reproductions of real objects, and image-to-3D conversion for transforming ideas or sketches into tangible objects. Regardless of the chosen method, it is essential to review the model before slicing and perform necessary adjustments in the slicer to ensure the best possible outcome.