CQ

Offset BUG

1

The paper has offset with shape [B, 2x9, H, W] But this repo has offset with shape [B, 2xC, H, W] I think this means two totally different things.

The number of parameters of this UNetRes is much larger than your previous work like DnCNN?

Questions about the number of training images

3

感谢您的工作和分享的code。 我刚开始了解去躁任务，想问一下训练集的选择和图片数目对结果有影响吗？因为我注意到dncnn那篇工作使用了不同的训练集。

Are you still planning on releasing training code?

I run with following script: ```sh python test_tracking.py --which_dataset KITTI --data_dir ./data/training/ --model_dir ./kitti_model/Pedestrian --model model.pth --category_name Pedestrian ``` And got following error: RuntimeError: Error(s) in loading state_dict for Pointnet_Tracking:...

No partpool provided.

2

No `/lip/parts_filter_done/`, can you provide the partpool images?

ICCV has already concluded, but there is still no code, right?

3

ICCV has already concluded, but there is still no code, right?

Question about sky

1

Could you provide more implementation details and reference literature about the `sky cubemap`?