08 2021 档案

Self-Supervised Visual Representations Learning by Contrastive Mask Prediction
摘要：一、解决什么样的问题？ 现有的基于contrastive learning的方法都有一个implicit semantic consistency(SC) assumption: a training image should have consistent semantic meaning acr 阅读全文
posted @ 2021-08-31 17:28 小孢子 阅读(394) 评论(0) 推荐(0)