Region-Based Semantic Image Clustering Using Positive and Negative Examples

Discovering various interest of users from massive image databases is a strenuous and rapid impel expedition region. Understanding the needs of users and representing them meaningfully is a challenging task. Region-based image retrieval (RBIR) is a method that incorporates the meaningful description of objects and an intuitive specification of spatial relationships. Our proposed model introduces a novel technique of semantic clustering in two stages. Initial semantic clusters are constructed in the first stage from the database log file by focusing on user interested query regions. These clusters are further refined by relevance feedback in the second stage based on probabilistic feature weight using positive and negative examples. Our results show that the proposed system enhances the performance of semantic clusters.

[1]  Viktor K. Prasanna,et al.  Semantic image clustering using object relation network , 2012, CVM.

[2]  Yannis Avrithis,et al.  Approximate Gaussian Mixtures for Large Scale Vocabularies , 2012, ECCV.

[3]  Shih-Fu Chang,et al.  VisualSEEk: a fully automated content-based image query system , 1997, MULTIMEDIA '96.

[4]  Zhiguo Gong,et al.  Web Image Semantic Clustering , 2005, OTM Conferences.

[5]  Hervé Jégou,et al.  Multiple Measurements and Joint Dimensionality Reduction for Large Scale Image Search with Short Vectors , 2015, ICMR.

[6]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[7]  Hervé Jégou,et al.  Orientation Covariant Aggregation of Local Descriptors with Embeddings , 2014, ECCV.

[8]  Florent Perronnin,et al.  Large-scale image retrieval with compressed Fisher vectors , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[9]  Djemel Ziou,et al.  Relevance feedback for CBIR: a new approach based on probabilistic feature weighting with positive and negative examples , 2006, IEEE Transactions on Image Processing.

[10]  Shin'ichi Satoh,et al.  Region-Based Image Retrieval Revisited , 2017, ACM Multimedia.

[11]  Michael Isard,et al.  Object retrieval with large vocabularies and fast spatial matching , 2007, 2007 IEEE Conference on Computer Vision and Pattern Recognition.

[12]  Ying Liu,et al.  Semantic Clustering for Region-Based Image Retrieval , 2007, Ninth IEEE International Symposium on Multimedia Workshops (ISMW 2007).

[13]  Cordelia Schmid,et al.  Aggregating Local Image Descriptors into Compact Codes , 2012, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[14]  Jiri Matas,et al.  Total recall II: Query expansion revisited , 2011, CVPR 2011.

[15]  Mingjing Li,et al.  Semantic image clustering using relevance feedback , 2003, Proceedings of the 2003 International Symposium on Circuits and Systems, 2003. ISCAS '03..

[16]  José Alfredo Ferreira Costa,et al.  The Semantic Clustering of Images and Its Relation with Low Level Color Features , 2008, 2008 IEEE International Conference on Semantic Computing.

[17]  Ming-Hsuan Yang,et al.  Context Driven Scene Parsing with Attention to Rare Classes , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[18]  Yiqiang Chen,et al.  Learning semantic cluster for image retrieval using association rule hypergraph partitioning , 2003, Fourth International Conference on Information, Communications and Signal Processing, 2003 and the Fourth Pacific Rim Conference on Multimedia. Proceedings of the 2003 Joint.

[19]  Jitendra Malik,et al.  Blobworld: Image Segmentation Using Expectation-Maximization and Its Application to Image Querying , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[20]  Hervé Jégou,et al.  Visual query expansion with or without geometry: Refining local descriptors by feature aggregation , 2014, Pattern Recognit..

[21]  Nuno Vasconcelos,et al.  Learning from User Feedback in Image Retrieval Systems , 1999, NIPS.

[22]  Jitendra Malik,et al.  Blobworld: A System for Region-Based Image Indexing and Retrieval , 1999, VISUAL.