Semi-supervised Learning with Constraints for Person Identification in Multimedia Data

被引：56

作者：

Baeuml, Martin ^{[1
]}

Tapaswi, Makarand ^{[1
]}

Stiefelhagen, Rainer ^{[1
]}

机构：

[1] Karlsruhe Inst Technol, D-76131 Karlsruhe, Germany

来源：

2013 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR) | 2013年

关键词：

D O I：

10.1109/CVPR.2013.462

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

We address the problem of person identification in TV series. We propose a unified learning framework for multi-class classification which incorporates labeled and unlabeled data, and constraints between pairs of features in the training. We apply the framework to train multinomial logistic regression classifiers for multi-class face recognition. The method is completely automatic, as the labeled data is obtained by tagging speaking faces using subtitles and fan transcripts of the videos. We demonstrate our approach on six episodes each of two diverse TV series and achieve state-of-the-art performance.

引用

页码：3602 / 3609

页数：8