Search results with tag "Contrastive learning"
Dense Contrastive Learning for Self-Supervised Visual Pre ...
openaccess.thecvf.com• We propose a new contrastive learning paradigm, i.e., dense contrastive learning, which performs dense pair-wise contrastive learning at the level of pixels (or local features). • With the proposed dense contrastive learning, we de-sign a simple and effective self-supervised learning method tailored for dense prediction tasks, termed
DetCo: Unsupervised Contrastive Learning for Object Detection
openaccess.thecvf.comIn this work, we present DetCo, which is a contrastive learning framework beneficial for instance-level detection tasks while maintaining competitive image classification transfer accuracy. DetCo contains (1) multi-level supervi-sion on features from different stages of the backbone net-work. (2) contrastive learning between global image and
An Empirical Study of Training Self-Supervised Vision ...
arxiv.orgSelf-supervised visual representation learning. In com-puter vision, contrastive learning [19] has become increas-ingly successful for self-supervised learning, e.g., [45,34, 22,2,20,10]. The methodology is to learn representa-tions that attract similar (positive) samples and dispel dif-ferent (negative) samples. The representations from con-