Table 11 Comparison of the fusion of two parts of features with and without transformer structure
From: BPI-MVQA: a bi-branch model for medical visual question answering
model | VQA-Med2018 | VQA-Med2019 | VQA-RAD | |||
|---|---|---|---|---|---|---|
W. | B. | A. | B. | A. | W. | |
BPI-MVQA without transformer fusion | 0.183 | 0.162 | 0.626 | 0.654 | 0.660 | 0.682 |
BPI-MVQA | 0.188 | 0.162 | 0.654 | 0.687 | 0.692 | 0.753 |