hi @jiepengwang
I am runniing the code on Scannet scene and I am observing the same problem as #8 , with layers of the same regions concatenated together in the final mesh. Could you please explain more in detail what is the solution to solve it?
I see one of your comments: "Hi, as illustrated in the paper, because of the SDF property, we can still get triangle faces at the invisible areas of the input images, when using marching cubes. You can remove the irrelevant faces by checking the visibility of the triangle faces to the input cameras. Or you can render depth maps then fuse them using tsdf fusion." Is this already implemented in the code, or it needs to be changed? Thank you!
hi @jiepengwang I am runniing the code on Scannet scene and I am observing the same problem as #8 , with layers of the same regions concatenated together in the final mesh. Could you please explain more in detail what is the solution to solve it? I see one of your comments: "Hi, as illustrated in the paper, because of the SDF property, we can still get triangle faces at the invisible areas of the input images, when using marching cubes. You can remove the irrelevant faces by checking the visibility of the triangle faces to the input cameras. Or you can render depth maps then fuse them using tsdf fusion." Is this already implemented in the code, or it needs to be changed? Thank you!