【cvpr2018】Solving Arbitrary Jigsaw Puzzles for Unsupervised Representation Learning(自监督)

本文介绍了一种从未标记图像数据中学习视觉特征的新方法,通过将图像划分为网格并打乱顺序,要求网络恢复原始配置,以此理解区域内和区域间的相互关系。实验表明,加入二元项可以提高拼图识别精度,并且3*3拼图优于2*2。尽管拼图识别准确率不高,但在迁移学习中表现优秀。
摘要由CSDN通过智能技术生成

解决问题

  • Learning visual features from unlabeled image data
  • a new approach in which the number of configurations can be arbitrarily large while the number of parameters remains
    unchanged.

 

相关工作

  • 聚类;
  • all images can be represented in a low-level space and trained encoders and/or decoders to recover the image and/or representation

 

Problem and Baseline Solution

  • an image is partitioned into a grid (e.g., 3*3) of patches and then disordered, and the task is to recover the original configuration. 
  • the network needs to understand what a patch contains as well as how two or more patches are related to each other
  • we expect this task to teach a network both intra-patch and inter-patch information

 

Strategy

  • 相对于9!的排列方式,作者提出新的算法
  • u
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值