adobe-research / layered-depth-refinement Goto Github PK

at the project page, I found the mask image is a binary mask that only have foreground and background, could you please tell me how to generate such type of mask ?

Hi, could you share some training source or tips? I want to add layered depth refinement to depth estimation task, but don't know how to start. Thanks

I read your wonderful paper, and believe it will be helpful for my research. But I enconter some issues when loding the maskdepth_model.pth checkpoint. Here is the error
Traceback (most recent call last): File "./test.py", line 104, in <module> load_ckpt(model, args.ckpt_path, device=device, strict=True) File "./test.py", line 52, in load_ckpt ckpt = torch.load(ckpt_path, map_location=lambda storage, loc: storage) File "/home/zzl/.conda/envs/zzl_torch_cuda_12/lib/python3.8/site-packages/torch/serialization.py", line 1028, in load return _legacy_load(opened_file, map_location, pickle_module, **pickle_load_args) File "/home/zzl/.conda/envs/zzl_torch_cuda_12/lib/python3.8/site-packages/torch/serialization.py", line 1246, in _legacy_load magic_number = pickle_module.load(f, **pickle_load_args) _pickle.UnpicklingError: invalid load key, 'v'.

Could you share your checkpoints?

ckpt/maskdepth_model.pth is 4.0K?

I'm getting

Error downloading object: ckpt/maskdepth_model.pth (a500b78): Smudge error: Error downloading ckpt/maskdepth_model.pth (a500b7897ee883d44f2f651e3dc2ba8de83bda5085877bf31fef30eefc3c0b2e): batch response: This repository is over its data quota. Account responsible for LFS bandwidth should purchase more data packs to restore access.

when trying to clone your repository

Thank you very much for your excellent work.

I met a problem loading ckpt from the pkl file you provide.

model = pickle.load(open(args.ckpt_path, 'rb'))

_pickle.UnpicklingError: A load persistent id instruction was encountered, but no persistent_load function was specified.

My environment is set as you said.