MediaPick: Tangible Semantic
Media Retrieval System
Gianpaolo D’Amico, Andrea Ferracani, Lea Landucci, Matteo Mancini,
Daniele Pezzatini and Nicola Torpei
Media Integration and Communication Center, University of Florence, Italy
Retrieval of information within a large amount of
multimedia data can be overwhelming.
Professionals who work daily with huge digital li-
braries (i.e. broadcast editorial staff) require ef-
ficient and ease to use interfaces to search, visu-
alize and organize video results collected in their
Users can browse concepts from an ontology
structure in order to select keywords and start
the video retrieval process.
The system exploits the search engine Sirio for
the retrieval and the semantic reasoning on mul-
The architecture consists of an infrared LED array
as an overlay built on top of an LCD screen. The
multi-touch overlay detects objects on its surface
and sends information using the TUIO protocol.
An input manager layer connects a socket server
to a gesture framework and a core logic, in or-
der to manage data from the video database.
ACM Multimedia 2010 - Demo Session - http://www.micc.unifi.it
email@example.com, firstname.lastname@example.org, email@example.com, firstname.lastname@example.org, email@example.com, firstname.lastname@example.org
MediaPick is a system for semantic multimedia
content based retrieval featuring a collaborative
natural interaction application.
The user interact with the system using gestures
on a multi-touch screen, performing ontology
based queries, retrieving semantic related infor-
mation and visually browsing several results.
The GUI consists of a concepts view, to select
keywords from an ontology structure to create the
query, and a results view, which shows the vid-
eos returned from the database in a results list.
Videos can be placed out of the list and moved
along the screen, resized or played. A cluster of
videos can be created by collecting two or more
Graphical user interface
Results listResults view