Anomaly detection has always been a critical and challenging problem in many application areas such as industry, healthcare, environment and finance. This problem becomes more di cult in the Big Data era as the data scale increases dramatically and the type of anomalies gets more complicated. In time sensitive applications like real time monitoring, data are often fed in streams and anomalies a...

In this paper, we propose an unsupervised video object cosegmentation framework based on the primary object proposals to extract the common foreground object(s) from a given video set. In addition to the objectness attributes and motion coherence our framework exploits the temporal consistency of the object-like regions between adjacent frames to enrich the set of original object proposals. We ...

In this paper, we introduce a secure transcoding framework that enables network intermediaries such as proxies to transcode multimedia data without violating end-to-end security guarantees. In our approach, an encoder decomposes a data stream at the source into multiple streams, encrypts each stream independently, and annotates each stream with clear-text metadata. An intermediary performs tran...

This paper introduces the use of mobile agents as the mechanism for policy enforcement in multi-agent multidomain systems. The focus is on the effective application of communication policies in the setup and maintenance of spanning data streams that cross multiple hosts in different domains. We have designed and implemented a mobile agent based framework (FlexFeed) that works in concert with th...

When multiple video streams share a wireless network, careful rate allocation is needed to prevent congestion, as well as to balance the video qualities among the competing streams. In this paper, we present a unified optimization framework for video rate allocation over wireless networks. Our framework applies to both unicast and multicast sessions, and accommodates both scalable and nonscalab...

The martingale framework for detecting changes in data stream, currently only applicable to labeled data, is extended here to unlabeled data using clustering concept. The one-pass incremental changedetection algorithm (i) does not require a sliding window on the data stream, (ii) does not require monitoring the performance of the clustering algorithm as data points are streaming, and (iii) work...

Data streaming applications are becoming more and more common due to the rapid development in the areas such as sensor networks, multimedia streaming, and on-line data mining, etc. These applications are often running in a decentralized, distributed environment. The requirements for processing large volumes of streaming data at real time have posed many great design challenges. It is critical t...

Savors is a visualization framework that supports the ingestion of data streams created by arbitrary command pipelines. Multiple data streams can be shown synchronized by time in the same or different views, which can be arranged in any layout. These capabilities combined with a powerful parallelization mechanism and interaction models already familiar to administrators allows Savors to display...