2016/05/03 by Andrea Emilio Salvi, Salvi, Andrea, Simone Ercoli +5
Computer Science · #Advanced Image and Video Retrieval Techniques #Caching and Content Delivery #FOS: Computer and information sciences #Information Retrieval (cs.IR) #Multimedia (cs.MM)
paper · pdf · doi:10.48550/arxiv.1605.00957
openalex publication_date 2016/05/03 · openalex created_date 2023/02/16 · openalex updated_date 2026/07/28
This paper presents a novel method for efficient image retrieval, based on a\nsimple and effective hashing of CNN features and the use of an indexing\nstructure based on Bloom filters. These filters are used as gatekeepers for the\ndatabase of image features, allowing to avoid to perform a query if the query\nfeatures are not stored in the database and speeding up the query process,\nwithout affecting retrieval performance. Thanks to the limited memory\nrequirements the system is suitable for mobile applications and distributed\ndatabases, associating each filter to a distributed portion of the database.\nExperimental validation has been performed on three standard image retrieval\ndatasets, outperforming state-of-the-art hashing methods in terms of precision,\nwhile the proposed indexing method obtains a 2\× speedup.\n