Partitioning Image Representation in Contrastive Learning. (arXiv:2203.10454v3 [cs.CV] UPDATED)

In contrastive learning in the image domain, the anchor and positive samples
are forced to have as close representations as possible. However, forcing the
two samples to have the same representation could be misleading because the
data augmentation techniques make the two samples different. In this paper, we
introduce a new representation, partitioned representation, which can learn
both common and unique features of the anchor and positive samples in
contrastive learning. The partitioned representation consists of two parts: the
content …

arxiv cv image learning partitioning representation