Jiaming Sun

Thanks for the reply. I tried to initialize weights with `tf.contrib.layers.variance_scaling_initializer(factor=2.0)`, which is the "MSRA-initialization" described in [this paper](https://arxiv.org/pdf/1502.01852.pdf), while it's not helping a lot. 1. What initialization did you...

You may refer to this https://github.com/lmb-freiburg/lmbspecialops/issues/3 and try to recompile tensorflow with a proper version of ABI.

Duplicate of #23. https://github.com/lmb-freiburg/demon/blob/devel/training/v2/training.py

BTW, it's better to change submodule `lmbspecialops @ 06fd1a4` to the latest commit to catch up with the changes in `flow_to_depth2`. What kind of bug have been fixed exactly in...

Thanks for your reply, for the multi GPU problem you may give me a guess on what could be wrong for me to start with if you don't have time...

Tried with a few things, but unfortunately with no outcomes either for a new build script for M1 or a usable COLMAP.app. I'm leaving my findings here: 1. As stated...

After some tweaking, I managed to get the code signing correct. [Here](https://mega.nz/file/8wxAnJJK#tcxS3zpb5u7cpJrK0GZ2nhiU4HlBQ6C9cyyEWjJBkxY) is the built COLMAP.app with native M1 binaries under the latest commit https://github.com/colmap/colmap/commit/30521f19de45c1cb2df8809728e780bf95fc8836 as of now. I will...

Hi @ShuLiu1993 , thanks for this wonderful work! I wonder if pretrained model on Cityscapes can be provided(or converted) in this PyTorch implementation? Thanks!

@ShuLiu1993 I can totally understand, thanks for your reply!

@ShuLiu1993 I'm trying to reproduce cityscapes training with COCO pretrained model, and I'm currently having 34.4 mask AP on Cityscapes validation set. I used the finetuing schedule described in the...