arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1803.03310cs.CV
收藏

度量学习中的泛化:嵌入层应该是嵌入层吗?

Generalization in Metric Learning: Should the Embedding Layer be the Embedding Layer?

Georgia Tech(佐治亚理工学院)

Nam Vo, James Hays

摘要:

这项工作研究了中小规模数据下的深度度量学习,因为我们相信更好的泛化可能是改进以前细粒度图像检索方法的一个因素;在设计未来技术时应该考虑到这一点。特别是,我们研究了在深度度量学习系统中使用其他层(除了嵌入层)进行特征提取,并分析了它们在训练数据上的表现,并推广到测试数据。从这项研究中,我们提出了一个新的正则化实践,其中可以添加或选择一个更优化的层特征提取。最先进的性能在3个细粒度图像检索基准测试中得到了证明:Cars-196,CUB-200-2011和Stanford Online Product。

英文摘要:

This work studies deep metric learning under small to medium scale data as we believe that better generalization could be a contributing factor to the improvement of previous fine-grained image retrieval methods; it should be considered when designing future techniques. In particular, we investigate using other layers in a deep metric learning system (besides the embedding layer) for feature extraction and analyze how well they perform on training data and generalize to testing data. From this study, we suggest a new regularization practice where one can add or choose a more optimal layer for feature extraction. State-of-the-art performance is demonstrated on 3 fine-grained image retrieval benchmarks: Cars-196, CUB-200-2011, and Stanford Online Product.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑