Batch Processing of Differently Shaped Tensor Data

• Project/Area Number: 23700212
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Perception information processing/Intelligent robotics
• Research Institution: Kyushu University
• Principal Investigator: INOUE Kohei 九州大学, 芸術工学研究科(研究院), 准教授 (70325570)
• Project Period (FY): 2011-04-28 – 2015-03-31
• Project Status: Completed (Fiscal Year 2014)
• Budget Amount: ¥3,770,000 (Direct Cost: ¥2,900,000、Indirect Cost: ¥870,000); Fiscal Year 2014: ¥780,000 (Direct Cost: ¥600,000、Indirect Cost: ¥180,000); Fiscal Year 2013: ¥780,000 (Direct Cost: ¥600,000、Indirect Cost: ¥180,000); Fiscal Year 2012: ¥780,000 (Direct Cost: ¥600,000、Indirect Cost: ¥180,000); Fiscal Year 2011: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000)
• Keywords: 画像情報処理 / テンソル / 行列 / 低ランク近似 / 一括処理 / 核ノルム / ベクトル / 次元削減

Outline of Final Research Achievements

For the purpose of the materialization of batch processing of differently shaped tensor data, several methods are proposed in this research includeing symmetric generalized low rank approximation of matrices, a non-iterative approximate solution method for 2DPCA, a dimensionality reduction method for vector data of different dimensions, simultaneous low rank approximation of matrices of different sizes, and simultaneous low rank approximation of matrices by nuclear norm minimization. The effectiveness of the proposed methods was experimentally demonstrated.

Research Products

• [Journal Article] 二次元主成分分析の非反復近似解法の一改良法 (2012)
  • Author(s): 井上光平, 原健二, 浦浜喜一
  • Journal Title: 電子情報通信学会論文誌 Volume: J95-D Pages: 1983-1984
  • NAID: 110009543899
  • Peer Reviewed
• [Journal Article] Non-iterative symmetric 2DLDA (2011)
  • Author(s): Kohei Inoue, Kenji Hara, and Kiichi Urahama
  • Journal Title: IEICE Trans. Information and Systems Volume: E94-D Pages: 926-929
  • Peer Reviewed
• [Presentation] 核ノルム最小化による行列データの同時低ランク近似 (2014)
  • Author(s): 井上光平, 原健二, 浦浜喜一
  • Organizer: 第29回信号処理シンポジウム
  • Place of Presentation: ビナリオ嵯峨嵐山
  • Year and Date: 2014-11-11 – 2014-11-14
• [Presentation] Memory-Efficient Computation of High-Dimensional Integral Images (2013)
  • Author(s): Kohei Inoue, Kenji Hara, Kiichi Urahama
  • Organizer: The 2nd IAPR Asian Conference on Pattern Recognition (ACPR2013)
  • Place of Presentation: Loisir Hotel & SPA Tower Naha
• [Presentation] サイズの異なる行列データの同時低ランク近似 (2013)
  • Author(s): 井上光平, 原健二, 浦浜喜一
  • Organizer: 第28回信号処理シンポジウム
  • Place of Presentation: 海峡メッセ下関
• [Presentation] A Hybrid Method for High Density Salt-and-Pepper Noise Removal (2013)
  • Author(s): Kohei Inoue, Kenji Hara, Kiichi Urahama
  • Organizer: The 9th International Conference on SIGNAL IMAGE TECHNOLOGY & INTERNET BASED SYSTEMS (SITIS2013)
  • Place of Presentation: Kyoto Terrsa Conference Center
• [Presentation] Integral Image-Based Error Diffusion (2013)
  • Author(s): Kohei Inoue, Kenji Hara, Kiichi Urahama
  • Organizer: Picture Coding Symposium (PCS2013)
  • Place of Presentation: San Jose, California
• [Presentation] 次元数の異なるベクトルデータの次元削減 (2012)
  • Author(s): 井上光平, 原健二, 浦浜喜一
  • Organizer: 第27回信号処理シンポジウム
  • Place of Presentation: ANAインターコンチネンタル石垣リゾート
• [Presentation] Symmetric GLRAM (2012)
  • Author(s): Kohei Inoue, Kenji Hara, and Kiichi Urahama
  • Organizer: ICASSP 2012
  • Place of Presentation: Kyoto International Conference Center
• [Presentation] 行列の一般化低ランク近似の対称化 (2011)
  • Author(s): 井上光平, 原健二, 浦浜喜一
  • Organizer: 第26回信号処理シンポジウム
  • Place of Presentation: 札幌コンベンションセンター
• [Presentation] 2次元主成分分析の非反復近似解法 (2011)
  • Author(s): 井上光平, 原健二, 浦浜喜一
  • Organizer: 2011年映像情報メディア学会冬季大会
  • Place of Presentation: 芝浦工業大学 豊洲キャンパス
• [Book] Applied Matrix and Tensor Variate Data Analysis (2015)
  • Author(s): Sakata, Toshio (Ed.)
  • Total Pages: 120
  • Publisher: Springer