feature_sampling problem

[Yvaine]

hi, thank you for your work, i have a problem in feature_sampling function, the following code reference_points_cam = (reference_points_cam - 0.5) * 2 why need multiply 2? Thank you for reply!

[zhanglazhuan]

for torch grid_sampler, the value should be [-1, 1] this code is trans [0, 1] to [-1, 1], I think so

[Yvaine]

for torch grid_sampler, the value should be [-1, 1] this code is trans [0, 1] to [-1, 1], I think so

Thank you~ That's correct.

[a1600012888]

Thanks, your understanding is correct.