I got the error when I submit my result: evalutils.exceptions.ValidationError: The images are not unique, please submit a unique image for each case.

I use team name as *.zip and the images in my zip file was equal to the proposed baseline_seg.zip file, but I still got the error.

Your zip file may have a subfold after unzipping.

There are only 101 segmentation result images in my result archive, no subfolders.

Since all the other submissions work well on the platform, it is probably the problem with your segmentation results.

Could you please evaluate your results with the official evaluation code? (using the baseline results as the pseudo "ground truth")

I evaluated my results as your said way(using the baseline results as the pseudo "ground truth"). It works. There is a F1 score after evaluating. I think your submissions evalting program may have some problem.

Each segmentation mask should have several cells with the label id 1,2,3,4,...

However, many of your segmentation results are empty.

a = tif.imread('cell_00063_label.tiff')

np.max(a) 0

Ok, I think you are right. Because the predict restults are analysied, and I found the the same reason as you said. Thank you!