This paper presents a mobile application that exploits interactive narrative storytelling through GPT-4 and a custom image generative pipeline to improve cultural tourism experiences. The application helps tourists visiting cities to program and personalize cultural city tours creating stories with the user as the protagonist. The app guides the users to choose Point-Of-Interests (POIs) and narrative genres of the narrative while the image generation pipeline provides them with visual and coherent representations of their actions in the story contributing to a more immersive and personalized experience. Technical challenges include producing coherent stories and real-time and quality images, maintaining visual composition and person identity, including multiple concepts, through prompt engineering. We validate the effectiveness of the application and the image generative pipeline through users studies which evaluate the educational potential of our approach.
Personalized Generative Storytelling with AI-Visual Illustrations for the Promotion of Knowledge in Cultural Heritage Tourism / Ferracani, Andrea; Bertini, Marco; Pala, Pietro; Nannotti, Gabriele; Principi, Filippo; Becchi, Giuseppe. - ELETTRONICO. - (2024), pp. 28-32. ( ACM Multimedia Workshop) [10.1145/3689094.3689465].
Personalized Generative Storytelling with AI-Visual Illustrations for the Promotion of Knowledge in Cultural Heritage Tourism
Ferracani, Andrea;Bertini, Marco;Pala, Pietro;Principi, Filippo;Becchi, Giuseppe
2024
Abstract
This paper presents a mobile application that exploits interactive narrative storytelling through GPT-4 and a custom image generative pipeline to improve cultural tourism experiences. The application helps tourists visiting cities to program and personalize cultural city tours creating stories with the user as the protagonist. The app guides the users to choose Point-Of-Interests (POIs) and narrative genres of the narrative while the image generation pipeline provides them with visual and coherent representations of their actions in the story contributing to a more immersive and personalized experience. Technical challenges include producing coherent stories and real-time and quality images, maintaining visual composition and person identity, including multiple concepts, through prompt engineering. We validate the effectiveness of the application and the image generative pipeline through users studies which evaluate the educational potential of our approach.| File | Dimensione | Formato | |
|---|---|---|---|
|
3689094.3689465.pdf
accesso aperto
Tipologia:
Pdf editoriale (Version of record)
Licenza:
Open Access
Dimensione
8.34 MB
Formato
Adobe PDF
|
8.34 MB | Adobe PDF |
I documenti in FLORE sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.



