Welcome to Stranger Sections a new style of challenge type on Xeek!
This challenge involves unsupervised segmentation of thin section images. There are 1000 unlabeled images in the dataset. Your job will be to provide a model that can accurately segment the blobs on the thin sections. You are free to explore any and all methods you see fit to effectively segment these images.
Hello! Once you choose to join the challenge, the 1000 images will be available for download on the Data tab. These images are split between train and test, but you are free to combine if you like. All images are unlabeled. It’s your job to segment the images using any method involving machine learning. Please see the challenge description for more details.
Thank you for reaching out to the Xeek Team. We do apologize but per the Xeek Challenge, our rules state that certain countries & regions are restricted from participating. Please refer to Section 2.b in our Terms of Use. You may visit that by clicking this Link.
After speaking with our Xeek team we would like to recommend that you access Xeek on a mobile connection to see this resolves the issue. If this does not solve the problem our team will escalate and work towards a solution.
Thanks for the question @swetap. Stranger Sections is an open challenge; anyone can submit a final piece of code for the challenge, and the Xeek Team will consider it for the prize pool. We wanted to see how many creative ideas the Xeek Community could generate. We will release a final leaderboard after the challenge has ended.
We are looking forward to seeing what you come up with.
Can we assume the image size will remain constant as it is for all testing and training data? If so, it will be much easier to code the padding in the network pre-processing section.
Hi, Yes through mobile is working! Interesting It was not working through 2 locations- home and work.
Best regards,
Jonas
P.S. I am not sure I I will have enough time for this competition. But in the future, we will know