Adding label-based supervised contrastive losses to text-to-image GANs reduces FID by up to 30.1% on COCO and improves IS on CUB across four baselines.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
A Framework For Image Synthesis Using Supervised Contrastive Learning
Adding label-based supervised contrastive losses to text-to-image GANs reduces FID by up to 30.1% on COCO and improves IS on CUB across four baselines.