Browse Prior Art Database

System and method to add semantic data to recordings files for interest based fragment retrieval Disclosure Number: IPCOM000199156D
Publication Date: 2010-Aug-27
Document File: 4 page(s) / 177K

Publishing Venue

The Prior Art Database


The article presents a system and a method to add semantic data to video recordings captured during web meetings.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 54% of the total text.

Page 1 of 4

System and method to add semantic data to recordings files for interest based fragment retrieval

Described below is one embodiment of the invention. Below is a description of the

annotations. The

                            can be associated with any moment of the timescale of the movie file. Its purpose is to add some more knowledge about what is presented at that moment in the movie.

The format of the simple annotations can be as follows:


simple plain text annotation

_annotation (description, timestamp)

description is a String (ex: "John started to speak about Video on Demand")

timestam a timestamp indicating the moment on the movie timescale for that annotation

The scope of a composed plain text annotation is to delimit a specific fragment of interest in the movie file. Basically, it will be a pair of two simple annotations like


plain text


_start, simple




[This page contains 1 picture or other non-text object]

Page 2 of 4

After the movie file of the recording is created it will be uploaded to the system. The proposed architecture of the system is a typical client-server design. The application server will host the Annotation System that is responsible with managing the recording files, offering the UI to annotate the files,

providing the UI to search in the collection

pause was issued. For a composed plain text annotation the user will need to

create both simple annotations, start and stop, only one of these will not be accepted by the system. After the user enters the annotation and submit it, the data contained will be persisted in another table along with the user id of the author of annotation.

of files. It will consist of the following modules:

1. Uploading application, responsible for the upload of the recording movie file to the System. Once uploaded, the file will be persisted in the Relational DB.
2. Plain text annotation application responsible for adding, updating, deleting simple or composed plain text annotation to any uploaded file in the system, searching and retrieving a fragment of interest within a movie file based on keywords search, searching and retrieving the list of files that matches a set of keywords.

3. RDF/XML editor responsible for UI friendly easily creation of RDF/XML annotations (ex: Protege editor)

4. SPARQL interpreter responsible for executing RDF/XML queries and providing the results of these queries.

The rel...