… One key difference to visual-only contrastive learning is on how we sample motion clips for
both motion-only and motion-visual learning. Instead of sampling randomly, we constrain to …

… We consider several state-of-the-art contrastive learning methods and demonstrate that CLTT
allows linear classification performance that approaches that of the fully supervised setting …

… entropy learning. In this paper, we tackle the above men… contrastive learning and make use
of contrastive learning for long… framework for contrastive learning of visual representations. In …

… contrastive learning framework and theoretically analyze the disadvantages of minimal sufficient
representation in contrastive learning, … for contrastive learning of visual representations. …

… not appear in visual-only learning. In this paper, we consider cross-modal contrastive learning
and … A simple framework for contrastive learning of visual representations. In International …

… alignment via contrastive learning to enhance the generalizability of learned visual
representations. … cross-attention strategy to explicitly learn the soft matching between local …

… In this paper, we propose a genuine group-level contrastive visual representation learning
… learning. Two mainstream unsupervised learning schemes are the instance-level contrastive …

… For a given anchor data point, contrastive losses attempt to make the distance between the
… of contrastive learning. Inspired by this approach, we propose a selfsupervised, contrastive …

… More recently, contrastive methods based on the Siamese structure achieve … for contrastive
visual representation learning. It significantly improves the quality of learned representations …

… video models were jointly trained via contrastive learning to learn joint cross-modal
representations. We build on this approach and compare models that learn directly from the spoken …