Participation policies:
1.     For participation in this challenge you should read and follow CC BY-NC-SA license
2.     Only semi-automatic methods are allowed (fully deployable with Docker container with
minimum human interaction required)
3.     Additionally to challenge dataset, participants may use publicly available pre-trained models
with indication of it and its source, however external private and public data is prohibited to be used.
4.     Performance results and summaries of all teams will be published on results page. Top 3 teams should make their code publicly available. Other than top 3 teams can choose whether to publish their code. Any team can publish paper with citation of challenge dataset paper.
5. Publication policy:

• All participants are recommended to write short method description paper. Description papers should follow the Springer Lecture Notes in Computer Science. 

• All authors of the each submitted method-description paper from participated teams are qualified as authors of conference paper.

• Any team may publish full paper only after embargo time. Paper should cite challenge dataset paper.

• Challenge organizers should publish challenge paper first (considered as embargo time)