高技术通讯2011，Vol.21Issue(12)：1312-1317,6.DOI:10.3772/j.issn.1002-0470.2011.12.015
基于多视角二维主动学习的多标签分类
Multi-label classification based on multi-view two-dimensional active learning
摘要
Abstract
This paper presents the multi-view two-dimensional active learning (MV-2DAL) algorithm for multi-label image classification so as to thoroughly explore the redundancies along the dimensions of sample, label and view, by the organic integration of the active learning with the multi-view learning. Taking a sample-label pair as the basic labeling unit, the algorithm calculates the uncertainties from the dimensions of sample and label within each view u-sing the two-dimensional active learning, and captures the uncertainties over different views based on the multi-view fusion. The overall uncertainty along the three dimensions is obtained to detect the most informative sample-label pairs. The experiments on the real-world multi-label image classification demonstrate that the proposed MV-2DAL algorithm is effective for redundancy reduction, and thus greatly alleviates the burden on human labeling.关键词
主动学习(AL)/多视角学习/多标签分类/图像分类/多模态融合Key words
active learning ( AL)/multi-view learning/multi-label classification/image classification/multi-model fusion引用本文复制引用
张晓宇..基于多视角二维主动学习的多标签分类[J].高技术通讯,2011,21(12):1312-1317,6.基金项目
中央级公益性科研院所基本科研业务费专项资金(ZD2011-7-3)和中国科学技术信息研究所科研预研基金(YY-201114)资助项目. （ZD2011-7-3）
