You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hello, great paper.
Just want to ask about the feature extraction speed.
I try to extract d2-net feature in Aachen day-night dataset.
However, it takes about around 20~30 secs per image, even with the multi-scale option is off.
I wonder is it normal to observe such phenomenon?
After tracing the code, it seems like the problem could be caused by the feature interpolation.
Do you have any comment on these things? Thanks.
The text was updated successfully, but these errors were encountered:
Hello! This is definitely not normal - the single-scale runtime on a GTX1080Ti is around 1 second and 3-4 seconds for multi-scale on a Tesla P100 16GB.
Are you running it on the GPU? If yes, what kind of GPU? What version of PyTorch / CUDA / cuDNN are you using? I know there are some issues with dilated convolutions in the latest versions of PyTorch (which is why I suggested to use the CUDA 8 PyTorch 1.0 release). However, I personally never had any issues with the feature interpolation.
Hello, great paper.
Just want to ask about the feature extraction speed.
I try to extract d2-net feature in Aachen day-night dataset.
However, it takes about around 20~30 secs per image, even with the multi-scale option is off.
I wonder is it normal to observe such phenomenon?
After tracing the code, it seems like the problem could be caused by the feature interpolation.
Do you have any comment on these things? Thanks.
The text was updated successfully, but these errors were encountered: