Make3D: Turn a 2D Picture Into a 3D Model

Stanford University service Make3D automatically converts a single picture into a 3D model. We covered similar technology from Fotowoosh in April 2007.

Make3D takes a two-dimensional image and creates a three-dimensional fly around model that includes depth and a range views. Photos can be uploaded directly or pulled into the site from Flickr.

The service is based on an algorithm created by Stanford students Andrew Ng, Ashutosh Saxena and Min Sun that won the best paper award at the 3D recognition and reconstruction workshop at the International Conference on Computer Vision in Rio de Janeiro in October 2007.

A January Stanford News Service piece explains it some more:

…the algorithm breaks the image up into tiny planes called “superpixels,” which are within the image and have very uniform color, brightness and other attributes. By looking at a superpixel in concert with its neighbors, analyzing changes such as gradations of texture, the algorithm makes a judgment about how far it is from the viewer and what its orientation in space is. Unlike some previous algorithms, the Stanford one can account for planes at any angle, not just horizontal or vertical. This allows it to create models for scenes that have planes at many orientations, such as the curved branches of trees or the slopes of mountains.

The service is in the same space as Microsoft Photosynth, but unlike Microsoft’s more extensive product that meshes many images together to create 3D models, Make3D is a one image only service. If you like, Make3D is Photosynth for the common man, quick, simple, and although the results don’t come close to Photosynth, they are still very impressive. A full gallery of Make3D renders can be found here.