TRANSPARENT DIGITAL INPUT

We spoke about digital to digital. We spoke about free or low cost scan apps for our devices. Now, someone needs to write an app that takes a scan of our analogue transparent drawing as a digital input, and then turns that into a three dimensional computer file.

I did a cursory look on the web, and I did not see where anyone is offering exactly this product.  (If anyone knows of this sort of app, can you let me know?)

We have had the capability to turn a series of photos into a 3D CAD model. A few years ago, Autodesk came out with 123D Catch. What you needed to do was input a series of photos of the building or object taken from all sides. That sounds familiar? It should as we also need to walk around to the back of the building to take photos so we can draw a complete three dimensional enclosure.

And there were other applications that let you create 3D CAD files from photos. GrabCad was / is an app in which you input data by drawing over a photograph. By essentially digitally tracing the outlines of the object on your photo, you are giving the basic 3D form. And then there were a series of sweeps or lofts so that you would use to flesh out your 3D form.  There is something called Sketch Fab, in which you input a series of photos taken all round the object, and then run it thru other applications to establish a point cloud.  I had never heard of a point cloud.

The great limitation of this type of data input is that you are dealing with opaque, representational images. Because there is no transparency, you either have to walk around the building or rotate the object, snapping all the while with your camera.  And then we leave it up to the computer to put it together, as it must be too much work for most humans.

But what if you input a transparent image? What if as input, you have already resolved the basic three dimensional form? And then what if the app was able to understand this transparent analogue input and then kick out for you a manipulable digital CAD file?

That would be really cool. And only because of transparent digital input would this work. You, the computational human, have resolved the three dimensional form. You, the thinking human have put this together in your head and have made a transparent drawing which demonstrates this. And then you are simply asking the computer, your assistant, to perform the mundane task of connecting the planes and lines. You don’t need a computer algorithm to perform this intellectual task, thank you very much.

So how about it? From my very simple transparent image above, it would seem to me to be straightforward for a computer algorithm to analyze which planes and lines are in front of which. And if that were to happen, then all of a sudden transparent drawing would be a critical tool of the digital age.