1 citations · 2 across the 5 of their papers we have counts for
1 paper · 1 filter
Risa Shinoda, Ryo Hayamizu, Kodai Nakashima +3
Pre-training is a strong strategy for enhancing visual models to efficiently train them with a limited number of labeled images. In semantic segmentation, creating annotation masks…