COLLABORATIVE IMAGE ANNOTATION USING IMAGE WEBS

The widespread availability of hand-held devices equipped with cameras has facilitated the creation of massive image collections. Our method links together image regions containing instances of the same object to form a graph called an Image Web. Such graphs represent relationships between images based on shared visual content. We demonstrate how to use Image Webs as conduits for symbolic information propagation among images. Symbolic information include annotations provided by users who perhaps have special expertise or were close to where the sensor data was captured. Such annotations can then be propagated to related images of the same object and benefit other users. Our algorithm gives similarity weights to edges in the Image Web graph. These weights are used to attenuate the relevance of annotations as they propagate along edges of the graph. Experiments show that our system supports multiple users to share images and annotate images collaboratively fast and accurately.