In this paper we present a method for still image memorability estimation. The proposed solution exploits feature maps extracted from two Convolutional Neural Networks pre-trained for object recognition and memorability estimation respectively. The feature maps are then enhanced using a soft attention mechanism in order to let the model focus on highly informative image regions for memorability estimation. Results achieved on a benchmark dataset demonstrate the effectiveness of the proposed method.
Leonardi, M., Celona, L., Napoletano, P., Bianco, S., Schettini, R., Manessi, F., et al. (2019). Image memorability using diverse visual features and soft attention. In Image Analysis and Processing – ICIAP 2019. 20th International Conference, Trento, Italy, September 9–13, 2019, Proceedings, Part II (pp.171-180). Springer [10.1007/978-3-030-30645-8_16].
Image memorability using diverse visual features and soft attention
Leonardi, Marco;Celona, Luigi
;Napoletano, Paolo;Bianco, Simone;Schettini, Raimondo;
2019
Abstract
In this paper we present a method for still image memorability estimation. The proposed solution exploits feature maps extracted from two Convolutional Neural Networks pre-trained for object recognition and memorability estimation respectively. The feature maps are then enhanced using a soft attention mechanism in order to let the model focus on highly informative image regions for memorability estimation. Results achieved on a benchmark dataset demonstrate the effectiveness of the proposed method.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
