Fig 1.
The method overview during training.
Firstly, the teacher was trained using the strong dataset (semantic annotated images) represented with blue lines. The teacher was then used to make semantic annotations for the weak dataset (bounding box annotated images). The student was then trained on both the pseudo-annotated dataset Dw′, represented by the orange lines, and the strong dataset Ds.
Table 1.
Tumor sizes of the three datasets.
Fig 2.
The single output (SO) Student is highlighted in colors whereas the decoder branch of the dual output (DO) Student is implicated in gray. For the DO Student the ouput of the two decoder branches are concatenated to form a dual-channeled output. The teachers have the same architecture as the SO Student, but with three downsamplings rather than four.
Table 2.
Teacher results.
Table 3.
Student results.
Table 4.
Scarcely trained teacher results.
Fig 3.
A sample of the results produced by the scarce students on the test set.
The figure shows the input image, bounding box, and ground truth (GT) mask in the three top rows, respectively. The baseline model, single output (SO) Student, and dual output (DO) Students corresponding outputs are shown in the three bottom rows, respectively.
Table 5.
Scarcely trained student results.