Human-computer and multimodal interaction are increasingly used in everyday life. Machines are able to get more from the surrounding world, assisting humans in different application areas. In this context, the correct processing and management of signals provided by the environments is determinant for structuring the data. Different sources and acquisition times can be exploited for improving recognition results. On the basis of these assumptions, we are proposing a multimodal system that exploits Allen’s temporal logic combined with a prevision method. The main object is to correlate user’s events with system’s reactions. After post-elaborating coming data from different signal sources (RGB images, depth maps, sounds, proximity sensors, etc.), the system is managing the correlations between recognition/detection results and events in real-time to create an interactive environment for the user. For increasing the recognition reliability, a predictive model is also associated with the proposed method. The modularity of the system grants a full dynamic development and upgrade with custom modules. Finally, a comparison with other similar systems is shown, underlining the high flexibility and robustness of the proposed event management method.
MIFTel: a multimodal interactive framework based on temporal logic rules / Marini, MARCO RAOUL. - (2019 Feb 28).
MIFTel: a multimodal interactive framework based on temporal logic rules
MARINI, MARCO RAOUL
28/02/2019
Abstract
Human-computer and multimodal interaction are increasingly used in everyday life. Machines are able to get more from the surrounding world, assisting humans in different application areas. In this context, the correct processing and management of signals provided by the environments is determinant for structuring the data. Different sources and acquisition times can be exploited for improving recognition results. On the basis of these assumptions, we are proposing a multimodal system that exploits Allen’s temporal logic combined with a prevision method. The main object is to correlate user’s events with system’s reactions. After post-elaborating coming data from different signal sources (RGB images, depth maps, sounds, proximity sensors, etc.), the system is managing the correlations between recognition/detection results and events in real-time to create an interactive environment for the user. For increasing the recognition reliability, a predictive model is also associated with the proposed method. The modularity of the system grants a full dynamic development and upgrade with custom modules. Finally, a comparison with other similar systems is shown, underlining the high flexibility and robustness of the proposed event management method.File | Dimensione | Formato | |
---|---|---|---|
Tesi_dottorato_Marini.pdf
accesso aperto
Tipologia:
Tesi di dottorato
Licenza:
Tutti i diritti riservati (All rights reserved)
Dimensione
1.55 MB
Formato
Adobe PDF
|
1.55 MB | Adobe PDF |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.