Cinematographic shot classification assigns a category to each shot on the basis of the field size, which is determined by the portion of the subject and of the environment shown in the field of view of the camera. This task is very important in the context of the creative field and can help freelancers in their daily activities when it is performed automatically. Novel and effective approaches capable of processing large volumes of images/videos and analyzing them effectively are becoming increasingly important. This paper presents a data-driven methodology to automatically classify cinematographic shots through deep learning techniques. In our study, we consider four classes of film shots: full figure, half figure, half torso and close up and we discuss three different scenarios in which the proposed work can be helpful. A new dataset of images was created to evaluate performances of the proposed methodology and to compare them with state-of-the-art techniques. Experimental results demonstrate the effectiveness of the proposed approach in performing the classification task with good accuracy.

Cinematographic Shot Classification through Deep Learning / Vacchetti, B.; Cerquitelli, T.; Antonino, R.. - (2020), pp. 345-350. ((Intervento presentato al convegno 44th IEEE Annual Computers, Software, and Applications Conference, COMPSAC 2020 tenutosi a esp nel 2020 [10.1109/COMPSAC48688.2020.0-222].

Cinematographic Shot Classification through Deep Learning

Vacchetti B.;Cerquitelli T.;
2020

Abstract

Cinematographic shot classification assigns a category to each shot on the basis of the field size, which is determined by the portion of the subject and of the environment shown in the field of view of the camera. This task is very important in the context of the creative field and can help freelancers in their daily activities when it is performed automatically. Novel and effective approaches capable of processing large volumes of images/videos and analyzing them effectively are becoming increasingly important. This paper presents a data-driven methodology to automatically classify cinematographic shots through deep learning techniques. In our study, we consider four classes of film shots: full figure, half figure, half torso and close up and we discuss three different scenarios in which the proposed work can be helpful. A new dataset of images was created to evaluate performances of the proposed methodology and to compare them with state-of-the-art techniques. Experimental results demonstrate the effectiveness of the proposed approach in performing the classification task with good accuracy.
978-1-7281-7303-0
File in questo prodotto:
File Dimensione Formato  
Cinematographic Shot Classification through Deep Learning.pdf

non disponibili

Tipologia: 2a Post-print versione editoriale / Version of Record
Licenza: Non Pubblico - Accesso privato/ristretto
Dimensione 284.99 kB
Formato Adobe PDF
284.99 kB Adobe PDF   Visualizza/Apri   Richiedi una copia
Pubblicazioni consigliate

Caricamento pubblicazioni consigliate

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/11583/2860675