Demos
Below are links to demonstration pages for some of our research
software. Each demo page contains a brief description of the application
together with guidelines on how to use it and a link to a stable
demonstration version of the application.
ANSES |
 |
|
"Watch the news while I was away and tell me what happened." This project combines a Video scene change algorithm, with the current text segmentation and summarization techniques to build an automatic news summarization and extraction system. Television broadcast news are captured both in Video/Audio format with the accompanying subtitles in text format. News stories are identified, extracted from the video, and summarized in a short paragraph which reduces the amount of information into a manageable size. Individual news video clips can be retrieved effectively by a combination of video and text, using a reversed indexed search engine to provide distilled information such as a summarized version of the orginal text and highlights important key words in the text. |
NewsRoom |
 |
|
NewsRoom is a modular platform for television broadcast news capture, automatic segmentation, indexing and browsing. The core of NewsRoom is its automatic story segmentation software. Its modular architecture allows to integrate various segmentation algorithms, organised in modules, such as a video scene detection software, lexical analysis algorithms, temporal feature analysis and high-level object-based audio/video feature extraction, such as optical character recognition and logotype detection. NewsRoom also provides a web-based interface for news browsing and searching, and lets users modify story segmentation, giving to NewsRoom a community-based "wiki" flavour. |
AETOS |
 |
|
This system demonstrates a novel interation technique to
support content-based image search in large image collections. Each
image is represented by a vertex in a directed graph. An arc is
established between two images if there are one or more features for
which the images are nearest neighbours. The arcs are weigted by the
proportion of the features for which the nearest neighbour relationship
holds. The resultant network exposes the sematic richness of the image
collection and allows the user to browse through this. |
VISHNU |
 |
|
Info Navigatior: This is a document search engine with several
new visualization front ends that aid navigation through the set of
documents returned by a query. Keywords from the document set are
identified and selected on the fly. These keywords are then used to
compute document clusters. The clusters can be displayed and explored
using three visualisations: Sammon cluster view, Dendro map or radial
interactive visualization. |
uBase |
 |
|
uBase: This application is a web-based image and video browser
for content-based retrieval. It integrates current research into
browsing and searching into a unified interface. The resulting browser
provides the user with the power to browse a large image collection
quickly and accurately. The browser seamlessly integrate the following
features:
|
