From what I heard, in Real images, if you submit to both Test and Challenge Test, then the submission with highest on "test-dev" split will be the final one.
How does it work on abstract scenes as there is no test-dev split?
Since, we do not have multiple splits for abstract dataset, your entry will be picked based on entire test set and you will be ranked according to entire test set, for which you are allowed only limited number of submissions.
Posted by: vqateam @ May 29, 2016, 12:28 p.m.