|Title / Titel||Tripod - TRI-Partite multimedia Object Description|
|Abstract (PDF, 14 KB)|
|Summary / Zusammenfassung||Tripod’s primary objective is to improve access to the enormous body of visual media. Applying an innovative multidisciplinary approach Tripod will utilise largely untapped but vast, accurate and regularly updated sources of semantic information to create innovative intuitive search services, enabling users to effortlessly and accurately
gain access to the image they seek from this ever expanding resource.
Any image tagged with a direction and location can be harnessed, either legacy photographs with place name captions or those produced by increasingly prevalent cameras with built in compass and GPS information. Tripod will augment images with spatial data to compute contextual information about the location and features of the
actual landscape pictured. Using 3D models, buildings and landscape features contained in the image will be identified and located within the picture. Methods from a number of disciplines will use the spatial data to derive additional semantic information about the area photographed. Techniques from Web search and text summarisation will be applied to automatically create textual descriptions of the photographs, producing a rich
readable and multifaceted caption far removed from merely location but encompassing culturally encoded notions such as socially connotated language of place such as suburb, west end, etc.
Like Google, Tripod is domain independent. Its services can be applied to a broad range of searching needs and languages due to the adaptability and flexibility of the caption generation. It is a solution for the exponential swell of images available as it will manage all images with a location. Tripod will empower the user by adapting the search infrastructure to their complex understanding and language. Thus Tripod manages access to images for the present and the future by bridging the semantic gap.
The main results (outputs) of the project will be
• A georeferenced image collection: a large collection of photographs with both location and direction information encoded in the images.
• Tools that automatically enhance (caption) georeferenced images by adding detailed text and conceptual information about the location of both the photograph and the photographer at the time the picture was taken.
• A spatial image search engine that allows users to locate images using a range of keyword, conceptual and spatial query constructors
|Project leadership and contacts /
Projektleitung und Kontakte
|Funding source(s) /
|Duration of Project / Projektdauer||Jan 2007 to Dec 2009|