Virtual embeddings and self-consistency for self-supervised learning

no code implementations • 13 Jun 2022 • Tariq Bdair, Hossam Abdelhamid, Nassir Navab, Shadi Albarqouni

We validate TriMix on eight benchmark datasets consisting of natural and medical images with an improvement of 2. 71% and 0. 41% better than the second-best models for both data types.

Data Augmentation Representation Learning +1