First steps towards photogrammetry in nuke!

I have some early results from playing with my super secret photogrammetry pipeline!

I’ve been comparing what it can do versus what is returned by 123d. These are built on a dataset from Dan Short from the American Museum of Natural History in NYC. Thanks Dan!

This is what I got back from 123d…before resizing for web it’s 4096 x 2048:

Mexico_Head_tex_0

And this is what I got back from my Secret Weapon(tm)…before resizing for web it’s 4096 x 4096:

color
I can actually choose whatever resolution I want the texture at….so any res is possible. I can probably increase it a bit because the original photos are pretty big.

Now take a look at these comparisons! The head on the left is 123d, the one on the right is the Secret Weapon(tm)….I intentionally deleted the geo behind the head on that one, so it’s absence isn’t the software’s fault. Notice the higher rez mesh (this can be higher or lower, depending on the adjustment of many, many settings…something you don’t get at all with 123d).

Screen Shot 2013-07-02 at 11.21.55 PM

Screen Shot 2013-07-02 at 11.21.51 PM

Screen Shot 2013-07-02 at 11.21.43 PM

Screen Shot 2013-07-02 at 11.21.38 PM

Screen Shot 2013-07-02 at 11.21.32 PM

Screen Shot 2013-07-02 at 11.21.32 PM

That’s just a regular texture mapped object though…if I can get the projections into nuke, that’ll be AWESOME…and I think I’m really close. Give me another day or two :-)

2 thoughts on “First steps towards photogrammetry in nuke!

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>