| http://www.muscle-noe.org/jobs.html
Description:
An unresolved, general problem is recognizing objects in images. We propose exploiting written language resources and web-based image mining for building a large-scale visual dictionary.
The project involves using text analysis and lexical resources to identify objects that might be found in a picture, and then constituting a large visual dictionary of those objects by trawling image repositories on the Web. This post-doc would be expected to produce a system that would construct an image ontology for tens of thousands of objects and comprising millions of images.
For example, from lexical resources or text mining, the post-doc's system might identify that an "English Toy Spaniel" is a type of dog. This fact would be automatically included in the ontology and then the system built by the researcher would automatically gather images of that animal. In a further step, these images would be used to create a signature for the image of an "English Toy Spaniel." In such a way, a very large image ontology would be created.
Research involves identifying portrayable objects in text, and extracting image signatures for each collection of objects. Such a project ideally requires a post-doc who has experience both with image processing and text mining. The benefits for multimedia understanding are vast, since we currently have no list of what objects can be found in an image, and no large representative sets of images for all these objects.
Description of the proposed joint programme of activities
The post-doc will spend the first nine-months at the CEA in France, studying the problem of building a large-scale image ontology. The definition of the ontology contents, and development of tools for mining the images for the ontology for the web would be completed at the CEA. As much of the ontology as possible would be instantiated at the CEA. Initial exploration of creating object signatures would be begun using the CEA image indexing tools.
The next nine-months would take this partially instantiated ontology to Bilkent where complete signatures would be built for all objects the ontology and experimentaton on object recognition using this ontology would be performed.
Candidate must currently be living in a country other than France or Turkey.
|
and GMD 