Martin Garbade

It was some time ago, but I think it worked in the end. The problem was the ['if clause'](https://github.com/shurans/sscnet/blob/master/matlab_code/perpareNYUCADdata.m#L119) in prepareNYUCADdata.m . You need to comment it out and always...

I remeber a similar repo I was using before and there the same spyware / virus discussions came up. Not that this is just copy / paste from that same...

Check out this modified version of [genSUNCGdataScript.m](https://github.com/mgarbade/sscnet/blob/master/matlab_code/genSUNCGdataScript.m#L46)

As far as I understand this the `slerp` code is only for interpolating between two points on a hypersphere. Isn't it also necessary to replace the z vector that is...

I've converted the data reading part of RLE files into boost python. Still uses GPU and C++ the code is simply interfaced to be accessible in python. You can check...

This is only a warning message, not an error. If you wait for a while the program is able to finish (takes ~1-2 min on my computer)

Same problem here. The convolutions of the underlying model are of the form ``` layer { name: "conv1_1" type: "Convolution" bottom: "data" top: "conv1_1" convolution_param { num_output: 16 pad: 3...

Any idea how to fix this? I mean one could add a slightly modified version in `Kaffe/tensorflow/network.py` which does respect 3D convolutions using [tf.nn.conv3d](https://www.tensorflow.org/versions/r0.12/api_docs/python/nn/convolution). Question is where in the code...

So after checking the code I'm quite sure that there is a missunderstanding on my side. At least the number of voxels that are randomly sampled from the empty occluded...

``` // Find number of occupied voxels // Save voxel indices of background // Set label weights of occupied voxels as 1 int num_occ_voxels = 0; std::vector bg_voxel_idx; float *occupancy_weight...