Table 3.

Results of segmentation on the test set for Task 1 by training the baseline on videos only from one center. Each center ID is also indicated (I - UCLH, II - IGG) for performance comparison between the two centers.

Train dataset	Video010	Video012	Video015	Video020	Video024	Video025	Overall mIoU
Center ID	II	II	I	I	II	I
I+II	0.5750	0.4122	0.6923	0.6757	0.5514	0.7045	0.6763
I	0.0109	0.0092	0.1012	0.0754	0.0056	0.2180	0.1102
II	0.1968	0.2630	0.1525	0.1562	0.3545	0.1907	0.1761