If you are a photographer (or at least enjoy taking pictures) and some sort of 3D texture artist, then you might be guilty of being “that guy (or girl) over there that is getting strangely excited about something on that concrete wall.” Now there is the potential to double that “strange excitement” with a new software program due to release anytime (at the time of this writing).

In the olden days, trickery was used to try to extract some sort of depth information (depth map, normal map) from a single photo. Programs like CrazyBump, xNormal*, and ShaderMap specialize in creating some or all of the following: Specular, Normal, Displacement and Ambient Occlusion maps. While theses programs may do a good job at creating these maps, they are limited to interpreting depth information from only one photo, like working with one eye. In which case, it would not be uncommon for more than a little preparation andor cleanup to be done in in a photo editing software such as Photoshop or Paint Shop Pro.

Now what if we could work with two eyes?

What if we could take two photos of an object or texture, separated by a few degrees, and have the depth information calculated from the parallax of the two images**? Enter PhotoSculpt Textures. Not only will it calculate more accurate depth information, it will also generate 3D geometry, Depth, Normal, Occlusion, and Specular maps. Created by an engineer over in France, PhotoSculpt Textures will be released soon (again, at the time of this writing) to the public. For downloadable samples, visit the PhotoSculpt Textures Gallery.

Without it, it makes you feel like a one-eyed pirate.

Here is a nice video showing off the whole process.

*xNormal provides a tool to create a Normal map from 4 photos, lit from 4 directions (Left, Right, Top, Bottom). Not really practical with anything more than small objects. Also, xNormal does many more cool things, as long as you can figure out how to use it.

**More information on Photogrammetry and Stereophotogrammetry here.