Image 2 (IMAGE)
Caption
Performance of the new approach under various label ratios on NIH ChestX-ray. Researchers denote the percentage of annotated training data in the target domain that REFERS needs to achieve comparable results with those of Model Genesis and ImageNet pre-training. Note that all three methods share the same transformer-based backbone.
Credit
The University of Hong Kong
Usage Restrictions
This work is licensed for non-commercial, non-exclusive, one-time usage with attribution to the copyright holder.
License
Original content