Closed jimmyfyx closed 5 months ago
We train the flow estimator to predict the object motion from
f_n
tof_{n-2}
as well asf_n
tof_{n+1}
(by picking one of them randomly).
Lines 64--75 essentially do that - pick one of them at random.
patch_size
is important. patch_size
should be bigger than the motion expected in the dataset. In other words, the cropped patches should have matching moving objects for the flow estimator to train. We choose the patch_size
by visual inspection.Thanks for the answer! Ah I see, sorry I missed your previous answer. But in the training data, shouldn't we need the flow from f_n
to f_{n-2}
and f_n
to f_{n+1}
at the same time? Why we need to pick one of them at random?
Flow estimator is trained separately, independent of what our entire framework is doing. The goal of this training is to learn flow estimation between MPIs. So, it doesn't matter which frames you estimate the flow between. The only thing that matters is that camera motion should be nullified between the frames.
Hi,
I have a few questions regarding the training part, especially the data format for training.
src/flow_estimation/data_loaders/VeedDynamic01.py
, within theload_training_data
function, what is the purpose of the block from line 64 to 70? Could you clarify its use a little bit?