Chop chop

[YoyoBoy]

I have a huge scene made of high-density Photoscan models mapped with huge bitmaps. At some point in the future I will need to port this to a game-engine-friendly world. I'm wondering if there is some sort of script that can chop up a model that is millions of polys into much smaller chunks, and re-do the mapping and UVs for each piece.

Or perhaps there are other ways to do this that I'm not aware of?

Thanks.

[Neilg]

the most automatic way would be a pro optimiser with keep textures and uv boundaries checked on, then using http://www.unwrella.com/ to try and do a better job unwrapping the now collapsed/vertex merged & optimised object then using render to texture. you can keep the high res object and bake some normals that way too.

[YoyoBoy]

ProOptimiser won't really work, since it's a large landscape created using drone footage (WIP screenshot attached), and at a human level, it's not that high-res.

Unwrella or Flatiron might do well with the unwrapping/baking part though.

[flino2004]

Rhino has several tools to do that if I understood well...If you get point cloud you could easy optimize it....reverse engineering tools may work too.
if you have the mesh you could create a point cloud of it ....the grid is depending of the level of detail that you want and from there you could have one big nurb of those point clouds.

[pg1]

Haven't tried this but you could use decimation master in zbrush and try one of its uv mapping modes, puv or auv tiles

[pg1]

Also have a look at the zremesher tool its pretty awesome I did a photoscan of my head and cleanup the topology really quickly.

[Joelaff]

You might also decimate it inside of Photoscan...