AlbertHuyb

l_dist term when instance_num=1

1

I am using you discriminative loss function for another use. And I discovered that if `instance_num==1`, the `l_dist` term will become `nan` because `mu_norm=[]`. So I suggest to add a...

Hello all, thanks for your nice work! I'm reading and learning about the code implementation for feature back projection. In [line 113 of Atlas/atlas/model.py](https://github.com/magicleap/Atlas/blob/bee3ddbdb0935cf9aa9fc810e829115c6ab74dc5/atlas/model.py#L113), the variable `self.origin` is defined as...

The basemodel in L10 and seg_opr in L11 of network.py are NOT FOUND in this repo. And the utils.py is also NOT FOUND. I can't reproduce your validation result, could...

[MERF] Per-scene results on mipnerf360 dataset?

2

Thanks for your nice work! Could you please release the per-scene results on the mipnerf360 dataset? Thanks! Best regards

How can we end-to-end train the MVSNet and Point Representation?

3

As mentioned in the paper, PointNeRF train the MVSNet for depth prediction and the Point Representation for rendering in an end-to-end manner. However, it seems that the training process in...

About the requirement of jax version.

1

I noticed that all versions of dm-pix library require `jax>=0.4.16`, even the `0.1.0` version. Is that necessary? I wonder if this can be modified so that I can use dm-pix...

i"m facing challenge with installing these. opencv-python==4.5.5.62

1

i"m facing challenge with installing these. opencv-python==4.5.5.62 git+https://github.com/zzd1992/Image-Local-Attention.git@master _Originally posted by @umarjibrilmohd in https://github.com/THU-LYJ-Lab/AR-Seg/issues/8#issuecomment-1779059146_

Thanks for your great work on this open source project!!! I noticed that README encourages us to run the GUI mode. Can you provide any instructions on the steps to...

About GPU memory consumption.

4

When training with `python train_chairs.py configs/train_chairs.yml`, I noticed that `batch_size=4` exceeds the memory limitation of a single 2080ti GPU. Instead, I can only set `batch_size` to `1` on a single...

About the training on FFHQ

8

Hello @Xuanmeng-Zhang , thanks for yoru excellent work. I want to confirm some details of training on FFHQ. 1. Did you train with 8 GPU cards? 2. How long does...