Interactive Search by Visual
Content
We
present image retrieval research activities at the IMEDIA group that are
integrated in our search engine IKONA based on client/server architecture. In
generic image databases, where image content is heterogeneous and no ground
truth is available or obvious, visual appearence is described by a combinatin
of general image signature such as color, texture and shape. We have developped
compact and efficient image signatures that relates spatial organisation of
color. Examples include stock photography and the World Wide Web. The user
should be assumed to be an average user (not an expert). In order to deal with
generic databases, Ikona includes a relevance feedback technique which enables
the user to refine their query by specifying over time a set of relevant and a
set of non-relevant images. Relevance feedback interaction methods try to use
the information the user supplies to the system in an attempt to
"guess" what are his intentions, thus making it easier to find what
he wants. It is an interactive way to minimize the semantic gap in such
low-level search by content. In specific image databases, we consider the available
ground truth and tune the models or range of parameters accordingly, maximising
the system efficiency. We have developped specific signature for face detection
and recognition and fingerprint identification. Region based queries are being
integrated into IKONA. In this mode, the user can select a part of an image and
the system will search images (or parts of images) that are visually similar to
the selected part. This ineteraction allow to the user to precise to the system
what part or particular object is interesting in the image. In this case, since
the query is focused, the system response is enhanced with regards to the user
target since the background image signature is not considered. We have
developed segmentation based methods as well as point of interest methods to
achive partial queries. While text indexing is ubiquitous, it is often limited,
tedious and subjective for describing image content. Visual content image
signatures are objective but has no semantic range. Combining both text and
image features for indexing and retieval is very promising area of interest of
IMEDIA team. We first work on a way to do keyword propagation based on visual
similarity. For example, if an image database has been partially annotated with
keywords, IKONA can suggest a number of keywords for a non annotated image and
their relevance. Further research on keyword propagation, semantic concept
search and hybrid text-image retrieval mode are being carried on. We will
present application with generic photo-gallery, specific face database and
criminal investigation department applications.