Skip to content

Experiment discussion #3

Open
Open

Description

@mathpluscode

Method

We agree to benchmark with https://github.com/voxelmorph/voxelmorph first.
But which paper/method should be tested first?

Comparison

We can compare

  • performances on different metrics (Dice, Hausdorff distance, SSD, etc.)
  • training time vs performance (using the same machine and train for the same amount of epochs)
  • results visualization

Activity

  1. YipengHu commented on Jan 15, 2021

    @YipengHu
    Member

    ok - i think if we can do these two above, it should be a good start. Three experiments, but with a bit more implementation for the probabilistic loss.

    I do not think we need exactly the same dataset as discussed.

    the unsupervised version applies mainly on atlas-based registration, so should be reasonably insensitive to pre-processing. #1

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions