Projects @ CMU Advanced Multimedia Processing Lab

Rate shaping selectively drops portions of the video bitstream before transmitting them in order to satisfy the network bandwidth requirement. In wireless multimedia transport over heterogeneous networks, the high error rate of the channel should be considered as well. We propose a rate shaping method that drops not only the source-coding segments of the video bitstream, but also the channel-coding segments of the video bitstream, adaptively according to the network condition

For more details on the project please see the Rate Shaping Project Page in AMP.

We introduce a new stochastic modeling technique called updating mixture of principal components (UMPC). UMPC specifically captures the non-stationary as well as the multi-modal characteristics of the data. Real-world data such as video data typically have these two characteristics. The video content changes over time and has a multi-modal probability distribution. We apply UMPC to perform error concealment for video data transmitted over networks with losses.

For more details on the project please see the Error Concealment Project Page in AMP.

We introduce a new stochastic process called the punctured autoregressive (AR) process, and use it to model both the variable bit rate (VBR) video traffic and the wireless channel dynamics. The model captures the long-range dependency (LRD) characteristics as well as the short-range dependency (SRD) characteristics of the video traffic and wireless channel dynamics.

For more details on the project please see the Traffic and Channel Modeling Project Page in AMP.

We propose a general framework for blind watermark detection. This framework contains a maximum-likelihood detector that utilizes the probability distribution of the original image. Other watermark detectors in literature are shown to be special cases of this framework. We demonstrate this framework in both the pixel domain and the transform domain, and show that our detector outperforms others because of 1) better modeling of the probability distribution of the original image, and 2) consideration to the human visual system in this framework.

For more details on the project please see the Watermark Detection Project Page in AMP.

Projects @ HP Cambridge Research Lab

Near-Similar Image Retrieval for Massive Database HP Research Lab Image Similarity Project Page

Projects @ HP's Cambridge Research Lab

Biometric Authentication CMU Secure Continuous Biometric-Enhanced Authentication Project Page