In the last years, solutions were proposed in the literature to alleviate the complexity of using sophisticated graphic suites for 3D scene generation by leveraging automatic tools. The most common approach based on the processing of text descriptions, however, may not represent the ideal solution, e.g., for fast prototyping purposes. This paper proposes an alternative methodology able to extract information about the objects and the layout of the scene to be created from a single 2D image. Compared to previous works, experimental results reported in this work show improvements in terms of similarity between the 2D and 3D scenes.
An automatic 3D scene generation pipeline based on a single 2D image / Cannavò, Alberto; Bardella, Christian; Semeraro, Lorenzo; De Lorenzis, Federico; Zang, Congyi; Jiang, Ying; Lamberti, Fabrizio. - STAMPA. - (2021), pp. 109-117. (Intervento presentato al convegno 8th International Conference on Augmented Reality, Virtual Reality and Computer Graphics (AVR 2021) tenutosi a Online nel September 7-10, 2021) [10.1007/978-3-030-87595-4_9].
An automatic 3D scene generation pipeline based on a single 2D image
Cannavò, Alberto;De Lorenzis, Federico;Lamberti, Fabrizio
2021
Abstract
In the last years, solutions were proposed in the literature to alleviate the complexity of using sophisticated graphic suites for 3D scene generation by leveraging automatic tools. The most common approach based on the processing of text descriptions, however, may not represent the ideal solution, e.g., for fast prototyping purposes. This paper proposes an alternative methodology able to extract information about the objects and the layout of the scene to be created from a single 2D image. Compared to previous works, experimental results reported in this work show improvements in terms of similarity between the 2D and 3D scenes.File | Dimensione | Formato | |
---|---|---|---|
Cannavò2021_Chapter_AnAutomatic3DSceneGenerationPi.pdf
accesso riservato
Tipologia:
2a Post-print versione editoriale / Version of Record
Licenza:
Non Pubblico - Accesso privato/ristretto
Dimensione
1.3 MB
Formato
Adobe PDF
|
1.3 MB | Adobe PDF | Visualizza/Apri Richiedi una copia |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/11583/2917644