TO BE DONE: Provide intro to tool section and discussing and
explaining the Categorization of important tools [by Nikolaos &
Giorgos]:
Currently
there are many tools that are used for multimedia annotation; the aim
of this
section is to give a brief description of the main tools
characteristics in
order to provide guidelines for their proper use. A basic
characteristic is the
types of content that a tool can annotate which are mainly images and
in the
case of M-Ontomat annotizer videos as well. Continuing another
important
characteristic of the annotation tools is the type of metadata they
use. The
main types of metadata are structural, descriptive and administrative.
Structural
metadata comment on the structure of the image and by some tools this
operation
is made automatically based on low level operations that are made on
the image.
Administrative metadata on the other hand include information
concerning the
creator the date etc and finally descriptive metadata comment on the
content of
the image. The format of metadata is also a characteristic of the
annotation tools
and the most common formats are OWL and RDF.
Additional characteristics of the
annotation tools concern the type of annotation. Hence some tools give
the
opportunity to the user to annotate an image using ontologies
while
others allow free annotation. Granularity is another such
characteristic that specifies
whether annotation is segment based or file based. Additionally a tool
may be threaded
or unthreaded. By threaded meaning the ability to respond or add to a
previous
annotation and to stagger/structure the presentation of annotations to
reflect
this. Ending additional tools characteristics are the access they
provide their
operation mode and whether they are open source or not.
Concluding the appropriateness
of a tool depends on the type of annotation a user
requires. The most used annotation tools were categorized
according to the characteristics described and also were used for
different types of annotations (according to the use cases) for a
discussion on the individual tools see
Semantic Web Image Annotation Tools .
1) Type of content: Jpeg Images,Video etc
2) Type of metadata: Descriptive, administrative, structural etc
3) Format of metadata: OWL, RDF
4) Annotation Level: Extraction of visual characteristics and
association with domain ontology concepts-operation using
ontologies.
5) Operation mode: plug-in, stand-alone
6) Open source: YES, NO
Suggestions by Jane:
7) Collaborative or individual
8) Granularity - file-based or segment-based (and sub-categories of types of segmentation)
9) Threaded or unthreaded (By threaded I mean the ability to
respond or add to a previous annotation and to stagger/structure
the presentation of annotations to reflect this.)
10) Access controlled or open access
See Semantic Web Image Annotation Tools for a discussion on the individual tools.