Open JannikLa opened 1 year ago
I haven't worked with the pytorch version that much especially recently, but just so I know -- when you say the MSE version worked, what were the hyperparameters that you used?
What do you mean by this line -- what does 'overfit' mean here?
If I set the initialization of the flow layer a lot higher (to 1 or 10) the network is able to overfit
Task (what are you trying to do/register?)
I am registering two T1 3D images from the OASIS data set (L2Reg challenge). The scans are preprocessed (skull-stripped, aligned, normalized). I am taking a fixed image from one subject and randomly select another subject as moving image (examples below).
I invested a lot if time in figuring out what is wrong. Maybe someone can give hints on what I could try or what might be wrong in my setup. Any help is appreciated!
What have you tried
I am using the pytroch implemention of voxelmorph. A dataloader loads the inputs to the network.
Details of experiments
self.flow.weight = nn.Parameter(Normal(0, 1e-5).sample(self.flow.weight.shape))
(networks.py, l 214)Image of training loss with maxed out smoothing in visualization ("baseline" setup from above)
Image of training loss (LNCC) when overfitting on a pair of images with large initialization weights (Normal(0, 10))
Image of input data