期刊
2017 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV)
卷 -, 期 -, 页码 2070-2079出版社
IEEE
DOI: 10.1109/ICCV.2017.226
关键词
-
资金
- Deep-Mind
We investigate methods for combining multiple self-supervised tasks-i.e., supervised tasks where data can be collected without manual labeling-in order to train a single visual representation. First, we provide an apples-to-apples comparison of four different self-supervised tasks using the very deep ResNet-101 architecture. We then combine tasks to jointly train a network. We also explore lasso regularization to encourage the network to factorize the information in its representation, and methods for harmonizing network inputs in order to learn a more unified representation. We evaluate all methods on ImageNet classification, PASCAL VOC detection, and NYU depth prediction. Our results show that deeper networks work better, and that combining tasks-even via a naive multihead architecture-always improves performance. Our best joint network nearly matches the PASCAL performance of a model pre-trained on ImageNet classification, and matches the ImageNet network on NYU depth prediction.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据