Dear @SimCol2022Participants,
One subtask of our challenge will be showing that your work generalizes to real data. To provide you with data from real colonoscopies, we partnered with the team behind the **EndoMapper** dataset!
Their dataset and more information can be found [here](https://www.synapse.org/#!Synapse:syn26707219/wiki/615178).
For the challenge, we will publish exclusive test sequences with COLMAP pseudo ground truth that will only be available in the scope of this challenge. However, this test data will be provided in the format as the EndoMapper data.
As their dataset is publicly available, you are allowed (and encouraged) to use the EndoMapper data for training! Please note that EndoMapper has its own data access regulations independent of ours. If you want to use their data for training purposes, please provide the keyword ?SimCol challenge? when you apply for access. Also, note that only some of the EndoMapper sequences provide COLMAP labels for training.
To summarize:
1. Go to https://www.synapse.org/#!Synapse:syn26707219/wiki/615178 and follow instructions on how to access the data **exactly**!
2. Update your Synapse user profile with your affiliation and Orcid information and state ?SimCol challenge? in your Statement of Intended Use when applying for access.
3. Note that some of the sequences have COLMAP labels. Our test data will be provided in the same format.
Thank you,
The SimCol organizers
P.S.: @SimCol2022preregistrants, if you are pre-registered but do not have access to the data yet, please get in touch with us via email!