Abstract : We address the problem of instance classification: our goal is to annotate images with tags corresponding to objects classes which exhibit small intra-class variations such as logos, products or landmarks. We propose a novel algorithm for the selection of class-specific prototypes which are used in a voting-based classification scheme. We show significant improvements over two state-of-the-art methods, namely the Fisher vector and Hamming Embedding, on two challenging methods of logos and vehicles.
Document type :
Conference papers
ICMR - ACM International Conference on Multimedia Retrieval, Apr 2014, Glasgow, United Kingdom. 2014
https://hal.inria.fr/hal-00942275
Contributor : Hervé Jégou
<>
Submitted on : Wednesday, February 5, 2014 - 11:56:17 AM
Last modification on : Thursday, July 19, 2018 - 12:16:04 PM
Document(s) archivé(s) le : Sunday, April 9, 2017 - 9:01:13 AM
