Task2 Evaluation Code (lungMasks vs. lms)

Task2 Evaluation Code (lungMasks vs. lms)  

  By: svesal on Aug. 16, 2021, 10:21 p.m.


I am trying to use the code for task_02 evaluation? however there is l2r_lms_loader that load a txt file. I could not understand why so? and what is the content of lms directory which is not part of training data that I have dwnloaded.

Best, Sulaiman

Re: Task2 Evaluation Code (lungMasks vs. lms)  

  By: AHering on Aug. 17, 2021, 7:12 p.m.

Hi Sulaiman,

the evaluation script is used for the evaluation of the test data and you don't have to use it during training. For task 2, we evaluate the target registration error of manually annotated landmarks. The landmarks are not available for the validation data, however, you can use the automatically generated keypoints (https://drive.google.com/file/d/1W2dvSrkZ7fodwEnxN8xbt4vQYgv3NHnb/view?usp=sharing) for a local evaluation if you like. You have to adapt the landmark loader a little.

Best, Alessa

Re: Task2 Evaluation Code (lungMasks vs. lms)  

  By: yuanpenggkate on May 12, 2023, 7:56 a.m.

Excuse me, the csv file in this keypoints compressed package has six columns, which generally represent the coordinates of the x, y, and z axes. What do the first three columns mean?

Re: Task2 Evaluation Code (lungMasks vs. lms)  

  By: chrgro on June 16, 2023, 5:49 p.m.

Dear yuanpenggkate, Please see my reply on your other post.

Best, Christoph