Skip to Main content Skip to Navigation
Conference papers

Memory Vectors for Particular Object Retrieval with Multiple Queries

Ronan Sicre 1 Hervé Jégou 1
1 LinkMedia - Creating and exploiting explicit links between multimedia fragments
Inria Rennes – Bretagne Atlantique , IRISA-D6 - MEDIA ET INTERACTIONS
Abstract : We address the problem of retrieving all the images containing a specific object in a large image collection, where the input query is given as a set of representative images of the object. This problem is referred to as multiple queries in the literature. For images described with bag-of-visual-words (BOW), one of the best performing approach amounts to simply averaging the query descriptors. This paper 1 introduces an improved fusion of the object description based on the recent concept of generalized max-pooling and memory vectors, which summarizes a set of vectors by a single representative vector. They have the property of reducing the influence of frequent features. Therefore , we propose to build a memory vector for each set of queries and the membership test is performed with each image descriptor from the database, to determine its similarity with the query representative. This new strategy for multiple queries brings a significant improvement for most of the image descriptors we have considered, in particular with Convolutional Neural Networks (CNN) features.
Document type :
Conference papers
Complete list of metadata

Cited literature [21 references]  Display  Hide  Download
Contributor : Teddy Furon Connect in order to contact the contributor
Submitted on : Wednesday, July 18, 2018 - 9:32:07 AM
Last modification on : Tuesday, October 19, 2021 - 11:04:39 AM
Long-term archiving on: : Friday, October 19, 2018 - 4:14:42 PM


Files produced by the author(s)



Ronan Sicre, Hervé Jégou. Memory Vectors for Particular Object Retrieval with Multiple Queries. ICMR 2018 - International Conference on Multimedia Retrieval, Jun 2015, Shanghai, China. pp.1-4, ⟨10.1145/2671188.2749306⟩. ⟨hal-01842224⟩



Record views


Files downloads