vidar icon indicating copy to clipboard operation
vidar copied to clipboard

Depthformer pre-trained checkpoint ?

Open tcourat opened this issue 2 years ago • 1 comments

Hi,

Where can I access the DepthFormer pre-trained checkpoint ? Also, is there a convenient way to use torchhub to load it like for the 3 other models (Define, Zerodepth and packnet) ? Having a little code snippet showing how to load DepthFormer and use it (expected input format) would help a lot. :)

Thanks

tcourat avatar Nov 23 '23 14:11 tcourat

zerodepth_model = torch.hub.load('/home/alance/.cache/torch/hub/TRI-ML_vidar_main', 'ZeroDepth', pretrained=True, source='local', path="/home/alance/.cache/torch/hub/checkpoints/ZeroDepth_unified.ckpt")

AlanceLau avatar Jan 19 '24 07:01 AlanceLau