This paper presents a novel deep architecture for saliency prediction. Current state of the art models for saliency prediction employ Fully Convolutional networks that perform a non-linear combination of features extracted from the last convolutional layer to predict saliency maps. We propose an architecture which, instead, combines features extracted at different levels of a Convolutional Neural Network (CNN). Our model is composed of three main blocks: a feature extraction CNN, a feature encoding network, that weights low and high level feature maps, and a prior learning network. We compare our solution with state of the art saliency models on two public benchmarks datasets. Results show that our model outperforms under all evaluation metrics on the SALICON dataset, which is currently the largest public dataset for saliency prediction, and achieves competitive results on the MIT300 benchmark.

A Deep Multi-Level Network for Saliency Prediction / Cornia, Marcella; Baraldi, Lorenzo; Serra, Giuseppe; Cucchiara, Rita. - (2016), pp. 3488-3493. (Intervento presentato al convegno 23rd International Conference on Pattern Recognition, ICPR 2016 tenutosi a Cancun, Mexico nel 4-8 Dec 2016) [10.1109/ICPR.2016.7900174].

A Deep Multi-Level Network for Saliency Prediction

CORNIA, MARCELLA;BARALDI, LORENZO;SERRA, GIUSEPPE;CUCCHIARA, Rita
2016

Abstract

This paper presents a novel deep architecture for saliency prediction. Current state of the art models for saliency prediction employ Fully Convolutional networks that perform a non-linear combination of features extracted from the last convolutional layer to predict saliency maps. We propose an architecture which, instead, combines features extracted at different levels of a Convolutional Neural Network (CNN). Our model is composed of three main blocks: a feature extraction CNN, a feature encoding network, that weights low and high level feature maps, and a prior learning network. We compare our solution with state of the art saliency models on two public benchmarks datasets. Results show that our model outperforms under all evaluation metrics on the SALICON dataset, which is currently the largest public dataset for saliency prediction, and achieves competitive results on the MIT300 benchmark.
2016
23rd International Conference on Pattern Recognition, ICPR 2016
Cancun, Mexico
4-8 Dec 2016
3488
3493
Cornia, Marcella; Baraldi, Lorenzo; Serra, Giuseppe; Cucchiara, Rita
A Deep Multi-Level Network for Saliency Prediction / Cornia, Marcella; Baraldi, Lorenzo; Serra, Giuseppe; Cucchiara, Rita. - (2016), pp. 3488-3493. (Intervento presentato al convegno 23rd International Conference on Pattern Recognition, ICPR 2016 tenutosi a Cancun, Mexico nel 4-8 Dec 2016) [10.1109/ICPR.2016.7900174].
File in questo prodotto:
File Dimensione Formato  
2016-icpr-saliency.pdf

Open access

Tipologia: Versione originale dell'autore proposta per la pubblicazione
Dimensione 4.54 MB
Formato Adobe PDF
4.54 MB Adobe PDF Visualizza/Apri
Pubblicazioni consigliate

Licenza Creative Commons
I metadati presenti in IRIS UNIMORE sono rilasciati con licenza Creative Commons CC0 1.0 Universal, mentre i file delle pubblicazioni sono rilasciati con licenza Attribuzione 4.0 Internazionale (CC BY 4.0), salvo diversa indicazione.
In caso di violazione di copyright, contattare Supporto Iris

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11380/1103794
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 296
  • ???jsp.display-item.citation.isi??? 224
social impact