Presentation | 2018-09-21 Codebook creation using agglomerative clustering for large scale image retrieval Kei Maeda, Toshikazu Wada, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Similar image retrieval can be applied to wide variety of applications including image forensic and landmark detection/recognition, and so on. Most of the methods employs nearest neighbor search among BoF vectors consisting of local image features. This approach has an advantage on scalability, but also have a drawback that the accuracy totally depends on the codebook. In this report, we assume that centers of big clusters consisting of varieties of local image features cause inaccurate retrieval. For solving this problem, we employ Ward’s method as a clustering algorithm, because it can generate almost uniform number of clusters. Also, we employ two-stage clustering, which merges clustering results for different images. This method is based on the observation that similar local image features are often involved in a single image. Through experiments, we confirmed our method can produce better retrieval accuracy than the methods proposed so far. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Similar image retrieval / BoF vector / clustering / Ward's method |
Paper # | PRMU2018-66,IBISML2018-43 |
Date of Issue | 2018-09-13 (PRMU, IBISML) |
Conference Information | |
Committee | PRMU / IBISML / IPSJ-CVIM |
---|---|
Conference Date | 2018/9/20(2days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | |
Chair | Shinichi Sato(NII) / Hisashi Kashima(Kyoto Univ.) |
Vice Chair | Yoshihisa Ijiri(Omron) / Toru Tamaki(Hiroshima Univ.) / Masashi Sugiyama(Univ. of Tokyo) / Koji Tsuda(Univ. of Tokyo) |
Secretary | Yoshihisa Ijiri(NEC) / Toru Tamaki(Osaka Univ.) / Masashi Sugiyama(Nagoya Inst. of Tech.) / Koji Tsuda(AIST) |
Assistant | Go Irie(NTT) / Yoshitaka Ushiku(Univ. of Tokyo) / Tomoharu Iwata(NTT) / Shigeyuki Oba(Kyoto Univ.) |
Paper Information | |
Registration To | Technical Committee on Pattern Recognition and Media Understanding / Technical Committee on Infomation-Based Induction Sciences and Machine Learning / Special Interest Group on Computer Vision and Image Media |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Codebook creation using agglomerative clustering for large scale image retrieval |
Sub Title (in English) | |
Keyword(1) | Similar image retrieval |
Keyword(2) | BoF vector |
Keyword(3) | clustering |
Keyword(4) | Ward's method |
1st Author's Name | Kei Maeda |
1st Author's Affiliation | Wakayama University(Wakayama Univ.) |
2nd Author's Name | Toshikazu Wada |
2nd Author's Affiliation | Wakayama University(Wakayama Univ.) |
Date | 2018-09-21 |
Paper # | PRMU2018-66,IBISML2018-43 |
Volume (vol) | vol.118 |
Number (no) | PRMU-219,IBISML-220 |
Page | pp.pp.213-218(PRMU), pp.213-218(IBISML), |
#Pages | 6 |
Date of Issue | 2018-09-13 (PRMU, IBISML) |