Sparse Isotropic Hashing

• Item type: Trans(1)
• 公開日: 2013-07-29
• タイトル: Sparse Isotropic Hashing
• タイトル / 言語: en
• タイトル: Sparse Isotropic Hashing
• 言語: eng
• キーワード / 主題Scheme: Other
• キーワード / 主題: [Regular Paper - Express Paper] binary codes, nearest neighbor search, local descriptors, sparse matrix
• 資源タイプ / 資源タイプ識別子: http://purl.org/coar/resource_type/c_6501
• 資源タイプ: journal article
• 著者所属: Denso IT Laboratory, Inc.
• 著者所属: Denso IT Laboratory, Inc.
• 著者所属: Denso IT Laboratory, Inc.
• 著者所属(英): en
• 著者所属(英): Denso IT Laboratory, Inc.
• 著者所属(英): en
• 著者所属(英): Denso IT Laboratory, Inc.
• 著者所属(英): en
• 著者所属(英): Denso IT Laboratory, Inc.
• 著者名: Ikuro, Sato; Mitsuru, Ambai; Koichiro, Suzuki
• 著者名(英): Ikuro, Sato; Mitsuru, Ambai; Koichiro, Suzuki
• 論文抄録 / 内容記述タイプ: Other
• 論文抄録 / 内容記述: This paper address the problem of binary coding of real vectors for efficient similarity computations. It has been argued that orthogonal transformation of center-subtracted vectors followed by sign function produces binary codes which well preserve similarities in the original space, especially when orthogonally transformed vectors have covariance matrix with equal diagonal elements. We propose a simple hashing algorithm that can orthogonally transform an arbitrary covariance matrix to the one with equal diagonal elements. We further expand this method to make the projection matrix sparse, which yield faster coding. It is demonstrated that proposed methods have comparable level of similarity preservation to the existing methods.
• 論文抄録(英) / 内容記述タイプ: Other
• 論文抄録(英) / 内容記述: This paper address the problem of binary coding of real vectors for efficient similarity computations. It has been argued that orthogonal transformation of center-subtracted vectors followed by sign function produces binary codes which well preserve similarities in the original space, especially when orthogonally transformed vectors have covariance matrix with equal diagonal elements. We propose a simple hashing algorithm that can orthogonally transform an arbitrary covariance matrix to the one with equal diagonal elements. We further expand this method to make the projection matrix sparse, which yield faster coding. It is demonstrated that proposed methods have comparable level of similarity preservation to the existing methods.
• 書誌情報: IPSJ Transactions on Computer Vision and Applications （CVA）; 巻 5, p. 40-44, 発行日 2013-07-29
• ISSN / 収録物識別子タイプ: ISSN
• ISSN / 収録物識別子: 1882-6695
• 出版者 / 言語: ja
• 出版者: 情報処理学会