cDc
most probably the example is not up to date with the latest code pls start from scene.mvs, hopefully that will work
Reoslution Level0 can be anything, it just means the original image is not downscaled so what is the original image resolution?
that explains it, the GPU run OOM in order to process such high resolution a divide and conquer approach needs to be implemented; I can help you with that, but...
no idea what PCD is, but sounds like a point cloud; a point cloud has no visibility info as by definition points do not have size; I suggest to use...
sure I do, pls share more details on what you have and what you need; ideally share more about the entire project so that I can make a better idea...
@pmoulon thx, I didn't read this exact paper yet (but an other one with a similar idea) @rjanvier the clustering is more for evenly partition the images in sub-sets, the...
pls note that I wrote SVM as the second step only to emphasize that it uses the sparse input and not the end meshes, but the computed boundaries are not...
The camera overlap is added so that the sub-set reconstructions have overlap meshes. However for the SVM boundary estimation non-overlaping sub-sets must be used. My mistake, I read the paper...
from my experience `SVM::C_SVC` plus `SVM::RBF` with auto-training should work fine and fast enough; the key is not having too many points to cluster; I suggest to use 2 points...