Outdoor Augmented Reality applications are an emerging class of software systems that demand the fast identification of natural objects on mobile and embedded systems. They arise as an interesting tool to support the creation of entertainment and educational applications. Artificial Intelligence has recently exhibited superior performance in a variety of Computer Vision tasks and can lead to novel Augmented Reality solutions. Nonetheless, their execution remains challenging and requires non negligible resources for devices with hardware constraints. The goal of the research presented in this thesis is to exploit the commoditization of Artificial Intelligence methods and the forthcoming wave of low-cost mass market Augmented Reality devices, to propose methods, architectures and components to support the creation and evaluation of solutions for outdoor Augmented Reality applications efficiently executable on low-power portable devices. Specifically, the focus is set on entertainment applications that can play a fundamental role to motivate citizens to contribute for environmental crowdsourcing purposes, such as data collection. The experimental results demonstrate how Artificial Intelligence, Computer Vision and Augmented Reality can be successfully integrated for the construction of novel entertaining solutions for limited-capacity portable systems.
Le applicazioni di realtà aumentata all'aperto sono una classe emergente di sistemi software che richiedono l'identificazione rapida di oggetti naturali su dispositivi mobili e sistemi embedded. Sono diventati sempre di più uno strumento interessante per supportare la creazione di applicazioni educative e di intrattenimento. L'intelligenza artificiale ha recentemente mostrato prestazioni superiori in diverse applicazioni di Computer Vision e può ora portare a nuove soluzioni in realtà aumentata. Tuttavia, la loro esecuzione rimane impegnativa e richiede risorse non trascurabili per i dispositivi con vincoli hardware. L'obiettivo della ricerca presentata in questa tesi è quello di sfruttare la mercificazione dei metodi di Intelligenza Artificiale e la prossima ondata di dispositivi di Realtà Aumentata del mercato di massa a basso costo, per proporre metodi, architetture e componenti per supportare la creazione e la valutazione di soluzioni per Applicazioni in Realtà Aumentata per le attività all’aria aperta in modo efficiente su dispositivi mobile a capacità limitata. In particolare, l'attenzione è rivolta alle applicazioni di intrattenimento che possono svolgere un ruolo fondamentale per motivare i cittadini a contribuire a scopi di crowdsourcing ambientale, come la raccolta di dati. I risultati sperimentali dimostrano come l'Intelligenza Artificiale, la Computer Vision e la Realtà Aumentata possano essere integrate con successo per la costruzione di nuove soluzioni divertenti per dispositivi mobile a capacità limitata.
Artificial Intelligence and Augmented Reality for entertainment applications
FRAJBERG, DARIAN
Abstract
Outdoor Augmented Reality applications are an emerging class of software systems that demand the fast identification of natural objects on mobile and embedded systems. They arise as an interesting tool to support the creation of entertainment and educational applications. Artificial Intelligence has recently exhibited superior performance in a variety of Computer Vision tasks and can lead to novel Augmented Reality solutions. Nonetheless, their execution remains challenging and requires non negligible resources for devices with hardware constraints. The goal of the research presented in this thesis is to exploit the commoditization of Artificial Intelligence methods and the forthcoming wave of low-cost mass market Augmented Reality devices, to propose methods, architectures and components to support the creation and evaluation of solutions for outdoor Augmented Reality applications efficiently executable on low-power portable devices. Specifically, the focus is set on entertainment applications that can play a fundamental role to motivate citizens to contribute for environmental crowdsourcing purposes, such as data collection. The experimental results demonstrate how Artificial Intelligence, Computer Vision and Augmented Reality can be successfully integrated for the construction of novel entertaining solutions for limited-capacity portable systems.File | Dimensione | Formato | |
---|---|---|---|
Darian_Frajberg_PhD_thesis.pdf
accessibile in internet per tutti
Descrizione: Darian Frajberg PhD thesis
Dimensione
29.03 MB
Formato
Adobe PDF
|
29.03 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/152139