Recently, I've taken an interest in the ability of photosynth to create a 3d point cloud from a collection of images. This allows for the basic photo to aggregate into a scaleless 3d digitizer. As an architecture student, I see the potentials in being able to easily switch between a physical and digital medium. Imagine generating a rhino or maya model from merely series of images.
Unfortunately, up until now, the 3d data is hidden from the general public. That is, until I stumbled across binarymillenium's awesome post on extracting and interpreting the data from photosynth. With Wireshark, a couple of batch scripts, and his python script, extracting the point clouds (.csv files) becomes a cinch. Thanks!
As for my part, I have written a simple rhinoscript that takes these data (.csv files) and enters them into rhino 4. I haven't gotten the colors to work yet. Still, this completes the circle of photos to 3d. With a little cleaning up and I've got myself a digital Duomo di Milano!
