There seems to be some bugs
GSeanCDAT opened this issue · 0 comments
GSeanCDAT commented
Hi,
Thanks for making your great work open-source.
When I come through your code, I find that there might be some bugs in the coorelation lookup function:
Lines 61 to 74 in 70f988f
It seems that the values of t1_scale and t0_scale show be switched as:
t1_scale = 1. / (1. - embt)
t0_scale = 1. / embt
Although it doesn't affect your presented results at fixed timestep 0.5, it might harm the model's performance when interpolating frames at an arbitrary time.