Gregory Kramida

The math in the paper is "incorrect", if I understood Slavcheva's email correctly. Their implementation has the "correct" math, and so does mine. In the bottom is the math for...

Hi, @linux-fan-dave. Sorry you are having trouble with pbcvt. I am not currently supporting embedded python development (I'm very far away from either embedded things or OpenCV at this point...

@KingfaLuis I'm not sure what you mean by "configure". The `configure` script of `autotools`? This project uses CMake.

Currently, there is already a simple thresholding filter that I implemented awhile back: one which checks if the average depth of surrounding neighbors of each pixel is or isn't within...

Tried in a separate branch to convert SDF units to voxels, it doesn't really work well, i.e. cannot get things to converge properly so far.

Keeping open for now -- might do more experiments later. So far it seems like scaling the SDF values would just scale the gradients, so I'm not sure how this...

Will need to probably update visible type in the fusion routine. Also, the previously allocated boundary hash blocks which received non-truncated values from the live frame will need to be...

The base initial implementation is done, but something about fusion doesn't seem to be working correctly.

Deferred until we're sure that the fusion works correctly.

Deferred until the Fusion step is confirmed to be working correctly across the first 20 frames.