search

sshan-zhao / ACMNet

Adaptive Context-Aware Multi-Modal Network for Depth Completion
64 stars 11 forks source link

clip option in test.py #12

Open dsshim0125 opened 1 year ago

dsshim0125 commented 1 year ago

Hi,

Thank you for your great work!

I am wondering why there is clip flag in run_test.sh.

With clip flag, the depth prediction results may become different from the original RGB and raw_velodyne as KITTI provided (prediction: 2561216, original RGB&sparse_depth: 3521216).

Is it ok to be different size of prediction and its input?

Thanks!