10
« on: January 18, 2019, 06:13:24 PM »
I need to build a super hi res mesh for a terrain surface from Mavic Pro drone dng images to be then baked as displacement/height /normal map textures of a shell low-poly one.
Looks like main limitation is RAM, right? Is there a way to work around it? I once build 800 mil mesh on the same pc but it's kind of foggy what settings I had that time.
I need maximum possible details in the target normal texture so I bet only way is to build a super huge mesh.
1. I should go "Heighfield" right? Not "Arbitrary" So the new depth map mode is not an option, right? I tired "depth map" and it stops at 97% decimation stage and never finishes it.
2. What should I set for "reuse depth map", does it matter at all?
3. I shouldn't even try "decimate" , right ? It doesn't work with huge meshes?
I actually somehow coped with "simplifing" my huge meshes in a Steam version of Reality Capture. Unfortunately RC itself fails with proper Mavic Pro images aligning and even when it doesn't, it does lots of surface artifacts over what Agisoft does and also inferior/ more blurry color textures IMO (Needs much more hi res input than the drone is able to provide considering flight/battery time)
4. What's best geometry format to export from Photoscan . Takes less time to save/ less GB on disc etc.? Alembic?
5. Would a color texture build for extra low poly mesh with custom unwrap be matching height/displacment details baked from super hi res mesh in 3d party texture baker? I mean considering texture bakers usually do ray tracing along low poly mesh normals? Would it match with how Agisoft projects the texture pixels?
Or I should make the color texture for same super hi poly object too? Would take forever I guess
ps. Could sombody who use drones share few thoughts. Is it good idea at al or anything I am going to get is lots of geometry noise instead of super hi res mesh?