fzy139

An alternative way is to download vgg19 manually from [vgg19.ckpt](http://download.tensorflow.org/models/vgg_19_2016_08_28.tar.gz) and modify the model path in .yml.

Acceleration is a good choice to predict the position indoor like the Inertial Navigation. Though its accuracy needs a lot optimization, many mobile AR devices use this method for indoor...

Looking forward to the release of full DMTet repo. Btw, does the current `kaolin.ops.conversions.marching_tetrahedra` implementation support extraction from the locally subdivided tet grid, as demonstrated in the paper? Thanks