Sparse representation based classification with intra-class variation dictionary on symmetric positive definite manifolds

Hiroyuki Kasai, Kohei Yoshikawa

研究成果: Conference contribution

抄録

Sparse representation based classification (SRC) using training samples as a dictionary has engendered promising results for many computer vision tasks. However, although the SRC classifier exhibits very competitive performances when given sufficient training samples of each class, it presents the difficulty that its performance decreases considerably when fewer training samples are used. As described herein, we propose a Riemannian SRC with intra-class variation dictionary on SPD matrices, R-ESRC. The key challenge is establishment of a mathematically correct intra-class variation dictionary in terms of geometry of SPD manifold. To this end, we exploit the geometric mean calculation and the logarithm mapping. Numerical evaluations demonstrate the superior performance of our proposed algorithm.

本文言語English
ホスト出版物のタイトル2017 IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2017
出版社Institute of Electrical and Electronics Engineers Inc.
ページ255-258
ページ数4
ISBN(電子版)9781538646625
DOI
出版ステータスPublished - 2018 6 18
外部発表はい
イベント17th IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2017 - Bilbao, Spain
継続期間: 2017 12 182017 12 20

出版物シリーズ

名前2017 IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2017

Conference

Conference17th IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2017
国/地域Spain
CityBilbao
Period17/12/1817/12/20

ASJC Scopus subject areas

  • 安全性、リスク、信頼性、品質管理
  • エネルギー工学および電力技術
  • コンピュータ ネットワークおよび通信
  • コンピュータ ビジョンおよびパターン認識
  • ハードウェアとアーキテクチャ
  • 信号処理

フィンガープリント

「Sparse representation based classification with intra-class variation dictionary on symmetric positive definite manifolds」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル