Q:  How long can the participation request be approved after sending the signed competition rule to NeurIPS_CellSeg@outlook.com?

A:  The request will be approved within 2 working days if the signed competition rule document is filled out correctly.

Q: I have registered for the competition but my requests have not been approved for more than two working days.

A: Please first make sure that you have filled out the online registration form (GoogleDoc or TencentDoc). Then, please reach out to `neurips.cellseg@gmail.com` and attach your registration information (name, grand-challenge id, and affiliation). We will double-check your requests.

Q: I'm only interested in the competition dataset but I do not want to join the competition. Can I download the dataset without joining the competition?

A: Thanks for your interest. To ensure enough submissions, the dataset is only available to participants during the competition. After the competition, we will make the dataset publicly available on Zenodo.

Q: How many people can form a team？

A: We do not have a limitation on the number of team members. 

Q: My team has multiple team members. Do they all need to join the grand-challenge?

A: No. We just require the team leader to join the grand-challenge. The team leader can share the competition data with the team members. Please note that all the team members should sign their names on the competition rule document. 

Q: I have registered for the competition and downloaded the dataset. Can I quit the competititon?

A: No! Please respect the signed agreement. If registered participants do not make successful submissions, all the team members will be listed on the competition website as registered but not completed and forbidden to join future challenges.  It will be available to the whole community which has a negative impact on your reputation.

Q: Can we use other datasets or pre-trained models to develop the segmentation algorithms?

A: External datasets and pre-trained models are allowed. However, you must post their links to the competition forum before 31st August, 2022. Private datasets and pre-trained models are not allowed for use.

Q: Can we manually annotate the unlabelled images to enlarge the training set?

A: No. Any manual interventions are not allowed. 

Q: Are there any limitations on the computational resources (e.g., GPU memory) for the training phase?

A: No. We only have a GPU memory constraint (<10 GB) in the testing phase.

Q: What are the expected outputs of the segmentation model?

A: This is an instance segmentation task. The output should assign a unique label for each cell (just like the ground truth in the training set).