Fine-grained vehicle classification is a challenging task due to the subtle differences between vehicle classes. Several successful approaches to fine-grained image classification rely on part-based models, where the image is classified according to discriminative object parts. Such approaches require however that parts in the training images be manually annotated, a labor-intensive process. We propose a convolutional architecture realizing a transform network capable of discovering the most discriminative parts of a vehicle at multiple scales. We experimentally show that our architecture outperforms a baseline reference if trained on class labels only, and performs closely to a reference based on a part-model if trained on loose vehicle localization bounding boxes.

Fine-Grained Vehicle Classification using Deep Residual Networks with Multiscale Attention Windows

Attilio Fiandrotti
Co-first
;
2017

Abstract

Fine-grained vehicle classification is a challenging task due to the subtle differences between vehicle classes. Several successful approaches to fine-grained image classification rely on part-based models, where the image is classified according to discriminative object parts. Such approaches require however that parts in the training images be manually annotated, a labor-intensive process. We propose a convolutional architecture realizing a transform network capable of discovering the most discriminative parts of a vehicle at multiple scales. We experimentally show that our architecture outperforms a baseline reference if trained on class labels only, and performs closely to a reference based on a part-model if trained on loose vehicle localization bounding boxes.
Multimedia Signal Processing (MMSP), 2017 IEEE 19th International Workshop on
Luton, United Kingdom
16-18 Ottobre 2017
Proceedings of the 2017 IEEE 19th International Workshop on Multimedia Signal Processing (MMSP)
IEEE
1
6
978-1-5090-3649-3
http://ieeexplore.ieee.org/document/8122262/
GHASSEMI, SINA; Attilio Fiandrotti; Enrico Magli; Gianluca Francini
File in questo prodotto:
File Dimensione Formato  
MMSP2017_P03_05_Ghassemi.pdf

Accesso riservato

Dimensione 964.52 kB
Formato Adobe PDF
964.52 kB 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: http://hdl.handle.net/2318/1770133
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 1
  • ???jsp.display-item.citation.isi??? 0
social impact