Multi-Modality Retrieval

Content-based Image Retrieval

img(Anaktisi) is a C#/.NET content base image retrieval application suitable for the web. It provides efficient retrieval services for various image databases using as a query a sample image, an image sketched by the user and keywords.

The image retrieval engine is powered by innovative compact and effective descriptors. Also, an Auto Relevance Feedback (ARF) technique is provided to the user. This technique readjusts the initial retrieval results based on user preferences improving the retrieval score significantly. img(Anaktisi) can be found at http://www.anaktisi.net.

  • Chatzichristofis, S.A. and Zagoris, K. and Boutalis, Y.S. and Papamarkos, N., “Accurate image retrieval based on compact composite descriptors and relevance feedback information”, International Journal of Pattern Recognition and Artificial Intelligence (IJPRAI), 24 (2), (207–244), 2010
  • Zagoris, K. and Chatzichristofis, S.A. and Papamarkos, N. and Boutalis, Y.S., “Automatic Image Annotation and Retrieval Using the Joint Composite Descriptor”, Informatics (PCI), 2010 14th Panhellenic Conference on, (143 -147), 2010
  • Zagoris, Konstantinos and Chatzichristofis, Savvas A. and Papamarkos, Nikos and Boutalis, Yiannis S., “img(Anaktisi): A Web Content Based Image Retrieval System”, Proceedings of the 2009 Second International Workshop on Similarity Search and Applications, (154–155), Prague, Czech Republic, 2009

Multimodal Retrieval Engine

An experimental multimodal search engine is presented at http://www.mmretrieval.net, which allows multimedia and multilingual queries in a single search and makes use of the total available information in a multimodal collection. All modalities are indexed separately and searched in parallel, and results can be fused with different methods depending on the noise and completeness characteristics of the modalities in a collection, whether the user is in a need of initial precision or high recall.

Beyond fusion, a 2-stage retrieval is provided by first thresholding the results obtained by secondary modalities, targeting recall, and then re-ranking the results based on the primary modality. The engine demonstrates the feasibility of the proposed architecture and methods on the ImageCLEF 2010 Wikipedia collection. The primary modality is image, consisting of 237434 items, associated with noisy and incomplete user-supplied textual annotations and the Wikipedia articles containing the images. Associated modalities are written in any combination of English, German, French, or any other unidentified language.

  • Arampatzis, Avi and Zagoris, Konstantinos and Chatzichristofis, Savvas A., “Dynamic two-stage image retrieval from large multimedia databases”, Information Processing & Management, 49 (1), (274 - 285), 2013
  • Arampatzis, Avi and Zagoris, Konstantinos and Chatzichristofis, Savvas A., “Fusion vs. two-stage for multimodal retrieval”, Proceedings of the 33rd European conference on Advances in information retrieval, (759–762), Dublin, Ireland, 2011
  • Arampatzis, Avi and Zagoris, Konstantinos and Chatzichristofis, Savvas A., “Dynamic two-stage image retrieval from large multimodal databases”, Proceedings of the 33rd European conference on Advances in information retrieval, (326–337), Dublin, Ireland, 2011
  • Zagoris, Konstantinos and Chatzichristofis, Savvas A. and Arampatzis, Avi, “Bag-of-visual-words vs global image descriptors on two-stage multimodal retrieval”, Proceedings of the 34th international ACM SIGIR conference on Research and development in Information, (1251–1252), Beijing, China, 2011
  • Zagoris, Konstantinos and Arampatzis, Avi and Chatzichristofis, Savvas A., “www.MMRetrieval.net: a multimodal search engine”, Proceedings of the Third International Conference on SImilarity Search and APplications, (117–118), Istanbul, Turkey, 2010
  • Avi Arampatzis and Savvas A. Chatzichristofis and Konstantinos Zagoris, “Multimedia Search with Noisy Modalities: Fusion and Multistage Retrieval”, CLEF (Notebook Papers/LABs/Workshops), 2010
☀️