arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1904.06925cs.CV
收藏

基于深度综合相关挖掘的图像聚类

Deep Comprehensive Correlation Mining for Image Clustering

SenseTime Research(商汤科技研究院);Peking University(北京大学);Shandong University(山东大学)

Jianlong Wu, Keyu Long, Fei Wang, Chen Qian, Cheng Li, Zhouchen Lin, Hongbin Zha

摘要:

最近开发的深度无监督方法允许我们联合学习表示和聚类未标记的数据。这些深度聚类方法主要关注样本之间的相关性,例如,选择高精度对以逐渐调整特征表示,这忽略了其他有用的相关性。本文提出了一种新的聚类框架--深度综合关联挖掘(DCCM),从三个方面探索和充分利用未标记数据背后的各种关联:1)提出伪标签监督来调查类别信息和学习区分特征,而不是只使用成对信息。2)充分挖掘了特征对输入空间图像变换的鲁棒性,有利于网络学习,显著提高了网络性能。3)针对聚类问题,提出了特征间的三元组互信息,将最近发现的实例级深度互信息提升到三元组层次,进一步帮助学习更具鉴别力的特征。在几个具有挑战性的数据集上进行的大量实验表明,我们的方法具有良好的性能,例如,在CIFAR-10上达到$62.3\%$聚类准确度,其比现有技术的结果高$10.1\%$。

英文摘要:

Recent developed deep unsupervised methods allow us to jointly learn representation and cluster unlabelled data. These deep clustering methods mainly focus on the correlation among samples, e.g., selecting high precision pairs to gradually tune the feature representation, which neglects other useful correlations. In this paper, we propose a novel clustering framework, named deep comprehensive correlation mining(DCCM), for exploring and taking full advantage of various kinds of correlations behind the unlabeled data from three aspects: 1) Instead of only using pair-wise information, pseudo-label supervision is proposed to investigate category information and learn discriminative features. 2) The features' robustness to image transformation of input space is fully explored, which benefits the network learning and significantly improves the performance. 3) The triplet mutual information among features is presented for clustering problem to lift the recently discovered instance-level deep mutual information to a triplet-level formation, which further helps to learn more discriminative features. Extensive experiments on several challenging datasets show that our method achieves good performance, e.g., attaining $62.3\%$ clustering accuracy on CIFAR-10, which is $10.1\%$ higher than the state-of-the-art results.

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。

↑