Fine-grained recognition focuses on the challenging task of automatically identifying the subtle differences between similar categories. Current state-of-the-art approaches require elaborated feature learning procedures, involving tuning several hyper-parameters, or rely on expensive human annotations such as objects or parts location. In this paper we propose a simple method for fine-grained recognition that exploits a nearly cost-free attention-based focus operation to construct an ensemble of increasingly specialized Convolutional Neural Networks. Our method achieves state-of-the-art results on three of the most popular datasets used for fine-grained classification namely CUB Birds 200-2011, FGVC-Aircraft and Stanford Cars requiring minimal hyper-parameter tuning and no annotations.

Increasingly Specialized Ensemble of Convolutional Neural Networks for Fine-Grained Recognition

Simonelli, Andrea;Stefano Messelodi;Samuel Rota Bulo
2018

Abstract

Fine-grained recognition focuses on the challenging task of automatically identifying the subtle differences between similar categories. Current state-of-the-art approaches require elaborated feature learning procedures, involving tuning several hyper-parameters, or rely on expensive human annotations such as objects or parts location. In this paper we propose a simple method for fine-grained recognition that exploits a nearly cost-free attention-based focus operation to construct an ensemble of increasingly specialized Convolutional Neural Networks. Our method achieves state-of-the-art results on three of the most popular datasets used for fine-grained classification namely CUB Birds 200-2011, FGVC-Aircraft and Stanford Cars requiring minimal hyper-parameter tuning and no annotations.
978-1-4799-7061-2
File in questo prodotto:
File Dimensione Formato  
increasingly-specialized-ensemble.pdf

solo utenti autorizzati

Tipologia: Documento in Pre-print
Licenza: NON PUBBLICO - Accesso privato/ristretto
Dimensione 2.31 MB
Formato Adobe PDF
2.31 MB Adobe PDF   Visualizza/Apri   Richiedi una copia

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11582/315951
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
social impact