The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Une méthode d'annotation basée sur la projection sur des ensembles convexes (POCS) pour la récupération d'images sémantiques est présentée dans cet article. En utilisant des images de base de données préalablement annotées par des mots-clés, la méthode proposée estime les caractéristiques sémantiques inconnues d'une image de requête à partir de ses caractéristiques visuelles connues, sur la base d'un algorithme POCS, qui comprend deux nouvelles approches. Premièrement, la méthode proposée attribue sémantiquement aux images de la base de données certains clusters et introduit un espace propre non linéaire de caractéristiques visuelles et sémantiques dans chaque cluster dans la contrainte de l'algorithme POCS. Cette approche fournit avec précision des caractéristiques sémantiques pour chaque cluster en utilisant ses caractéristiques visuelles au sens des moindres carrés. De plus, la méthode proposée surveille l'erreur convergente par l'algorithme POCS afin de sélectionner le cluster optimal incluant l'image de requête. En introduisant les deux approches ci-dessus dans l'algorithme POCS, les caractéristiques sémantiques inconnues de l'image de requête sont estimées avec succès à partir de ses caractéristiques visuelles connues. Par conséquent, des images similaires peuvent être facilement récupérées à partir de la base de données sur la base des caractéristiques sémantiques obtenues. Les résultats expérimentaux vérifient l'efficacité de la méthode proposée pour la récupération d'images sémantiques.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copier
Takahiro OGAWA, Miki HASEYAMA, "POCS-Based Annotation Method Using Kernel PCA for Semantic Image Retrieval" in IEICE TRANSACTIONS on Fundamentals,
vol. E91-A, no. 8, pp. 1915-1923, August 2008, doi: 10.1093/ietfec/e91-a.8.1915.
Abstract: A projection onto convex sets (POCS)-based annotation method for semantic image retrieval is presented in this paper. Utilizing database images previously annotated by keywords, the proposed method estimates unknown semantic features of a query image from its known visual features based on a POCS algorithm, which includes two novel approaches. First, the proposed method semantically assigns database images some clusters and introduces a nonlinear eigenspace of visual and semantic features in each cluster into the constraint of the POCS algorithm. This approach accurately provides semantic features for each cluster by using its visual features in the least squares sense. Furthermore, the proposed method monitors the error converged by the POCS algorithm in order to select the optimal cluster including the query image. By introducing the above two approaches into the POCS algorithm, the unknown semantic features of the query image are successfully estimated from its known visual features. Consequently, similar images can be easily retrieved from the database based on the obtained semantic features. Experimental results verify the effectiveness of the proposed method for semantic image retrieval.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1093/ietfec/e91-a.8.1915/_p
Copier
@ARTICLE{e91-a_8_1915,
author={Takahiro OGAWA, Miki HASEYAMA, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={POCS-Based Annotation Method Using Kernel PCA for Semantic Image Retrieval},
year={2008},
volume={E91-A},
number={8},
pages={1915-1923},
abstract={A projection onto convex sets (POCS)-based annotation method for semantic image retrieval is presented in this paper. Utilizing database images previously annotated by keywords, the proposed method estimates unknown semantic features of a query image from its known visual features based on a POCS algorithm, which includes two novel approaches. First, the proposed method semantically assigns database images some clusters and introduces a nonlinear eigenspace of visual and semantic features in each cluster into the constraint of the POCS algorithm. This approach accurately provides semantic features for each cluster by using its visual features in the least squares sense. Furthermore, the proposed method monitors the error converged by the POCS algorithm in order to select the optimal cluster including the query image. By introducing the above two approaches into the POCS algorithm, the unknown semantic features of the query image are successfully estimated from its known visual features. Consequently, similar images can be easily retrieved from the database based on the obtained semantic features. Experimental results verify the effectiveness of the proposed method for semantic image retrieval.},
keywords={},
doi={10.1093/ietfec/e91-a.8.1915},
ISSN={1745-1337},
month={August},}
Copier
TY - JOUR
TI - POCS-Based Annotation Method Using Kernel PCA for Semantic Image Retrieval
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 1915
EP - 1923
AU - Takahiro OGAWA
AU - Miki HASEYAMA
PY - 2008
DO - 10.1093/ietfec/e91-a.8.1915
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E91-A
IS - 8
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - August 2008
AB - A projection onto convex sets (POCS)-based annotation method for semantic image retrieval is presented in this paper. Utilizing database images previously annotated by keywords, the proposed method estimates unknown semantic features of a query image from its known visual features based on a POCS algorithm, which includes two novel approaches. First, the proposed method semantically assigns database images some clusters and introduces a nonlinear eigenspace of visual and semantic features in each cluster into the constraint of the POCS algorithm. This approach accurately provides semantic features for each cluster by using its visual features in the least squares sense. Furthermore, the proposed method monitors the error converged by the POCS algorithm in order to select the optimal cluster including the query image. By introducing the above two approaches into the POCS algorithm, the unknown semantic features of the query image are successfully estimated from its known visual features. Consequently, similar images can be easily retrieved from the database based on the obtained semantic features. Experimental results verify the effectiveness of the proposed method for semantic image retrieval.
ER -