In our prior work, we introduced a generalization of the multiple-instance learning (MIL) model in which a bag's label is not based on a single instance's proximity to a...
In this paper, a novel data hiding technique is proposed, as an improvement over the Fibonacci LSB data-hiding technique proposed by Battisti et al. [1] based on decomposition of ...
—Wireless Multimedia Sensor Networks (WMSNs) are distributed systems of wirelessly networked devices that allow retrieving video and audio streams, still images, and scalar senso...
Multiple-instance learning (MIL) is a generalization of the supervised learning problem where each training observation is a labeled bag of unlabeled instances. Several supervised ...
The phenomenal growth of video on the web and the increasing sparseness of meta information associated with it forces us to look for signals from the video content for search/info...
Ming Zhao 0003, Jay Yagnik, Hartwig Adam, David Ba...