search

alexklwong / mondi-python

PyTorch Implementation of Monitored Distillation for Positive Congruent Depth Completion (ECCV 2022)
44 stars 3 forks source link

Number of valid pixels problem #6

Closed KyuminHwang closed 1 year ago

KyuminHwang commented 1 year ago

When taking void_test_sparse_depth_1500 of the VOID dataset, there are cases where there are less than 1500 sparse points greater than 0. Have you ever checked it?

Even at 150pixel and 500pixel, the number of valid pixels does not satisfy the suggested 150, 500.

Thank you and look forward to your reply.

alexklwong commented 1 year ago

Yes that's correct, 1500, 500, 150 are the highest number of points. VOID is a realistic VIO dataset, where the number of tracked points are not guaranteed to be exactly 1500, 500, 150.

KyuminHwang commented 1 year ago

Thank you for your quick and kind reply :) Then I'll close the issue.