Skip to Main content Skip to Navigation
New interface
Reports (Research report)

Using extreme value theory for image detection

Teddy Furon 1 Hervé Jégou 1 
1 TEXMEX - Multimedia content-based indexing
IRISA - Institut de Recherche en Informatique et Systèmes Aléatoires, Inria Rennes – Bretagne Atlantique
Abstract : The primary target of content based image retrieval is to return a list of images that are most similar to a query image. This is usually done by ordering the images based on a similarity score. In most state-of-the-art systems, the magnitude of this score is very different from a query to another. This prevents us from making a proper decision about the correctness of the returned images. This paper considers the applications where a confidence measurement is required, such as in copy detection or when a re-ranking stage is applied on a short-list such as geometrical verification. For this purpose, we formulate image search as an outlier detection problem, and propose a framework derived from extreme values theory. We translate the raw similarity score returned by the system into a relevance score related to the probability that a raw score deviates from the estimated model of scores of random images. The method produces a relevance score which is normalized in the sense that it is more consistent across queries. Experiments performed on several popular image retrieval benchmarks and state-of-the-art image representations show the interest of our approach.
Document type :
Reports (Research report)
Complete list of metadata

Cited literature [22 references]  Display  Hide  Download
Contributor : Hervé Jégou Connect in order to contact the contributor
Submitted on : Sunday, June 29, 2014 - 10:25:30 PM
Last modification on : Thursday, October 27, 2022 - 3:45:22 AM
Long-term archiving on: : Tuesday, April 11, 2017 - 9:18:48 AM


Files produced by the author(s)


  • HAL Id : hal-00789804, version 2


Teddy Furon, Hervé Jégou. Using extreme value theory for image detection. [Research Report] RR-8244, INRIA. 2013. ⟨hal-00789804v2⟩



Record views


Files downloads