This dissertation investigates the modeling and optimization of several problems that arise durin... more This dissertation investigates the modeling and optimization of several problems that arise during compression or adaptation of both frame-based and object-based (structured) multimedia content. In particular, we address: (a) The optimal dynamic rate shaping (DRS) of Markov-1 sources. We present the first extensive experimental study on the various DRS algorithms (causally optimal, memoryless, and rate-based) both in their constrained and generalized forms. The study proves the computational viability of the DRS approach to transcoding and identifies a range of rate shaping ratios for which it is better than requantization, both complexity-wise as well as in performance. We substantiate the almost-optimal experimental performance of the memoryless algorithm by analyzing the behavior of the DRS problem assuming a first order Markovian source. (b) The modeling of object based audio-visual communication systems, such as MPEG-4, both at the server ( proxy) side for transcoding purposes ...
Proceedings 2003 International Conference on Image Processing (Cat. No.03CH37429)
We discuss the behavior of the optimal solution to the dynamic rate shaping problem assuming an A... more We discuss the behavior of the optimal solution to the dynamic rate shaping problem assuming an AR(1) source model. By analyzing the statistical and rate-distortion behavior of the different components of this minimization problem, the following key result is mathematically proven: "The set of optimal breakpoint values for any frame is invariant to the accumulated motion compensated shaping error from past frames and may be very reasonably approximated using the current frame shaping error alone".
IEEE International Conference on Acoustics Speech and Signal Processing, 2002
We discuss formulations to solve the bit-allocation problem with dependencies. These formulations... more We discuss formulations to solve the bit-allocation problem with dependencies. These formulations are more generic than those known earlier in that we relax a so-called causality assumption. We study the incremental improvement obtained by having a dependency structure in optimal quantization choices. As a way to solve this problem approximately, we discuss novel rounding procedures based on some relaxation of the original problem (either the LP relaxation or a Lagrange relaxation). The rounding algorithms perform extremely close to optimality. Our results also show an improvement of around 0.5 dB over the best causally-optimal solution.
We i n troduce the concept of Dynamic Rate Shaping DRS, a technique to adapt the rate of compress... more We i n troduce the concept of Dynamic Rate Shaping DRS, a technique to adapt the rate of compressed video bitstreams e.g., MPEG-1, MPEG-2, H.261, JPEG to dynamically varying rate constraints by operating directly in the compressed signal domain. Such a s c heme is shown to be critical for multimedia communication systems, since it can guarantee universal interoperability b e t ween encoders decoders and networks with widely di erent and even time-varying quality of service guarantees. The concept is shown to evolve naturally by considering rst the operation of an actual multimedia communication system, in the form of the Xphone" testbed that we h a ve developed. It is shown that a key technique for enabling video communication using a system based on a best-e ort operating system and network is, among others, adaptive rate control of the bit rate of video JPEG in this case. A natural extension of the adaptive rate control approach in Xphone is Data Partitioning. This scheme splits a compressed bitstream into two parts, and achieves robustness by transmitting them over channels with di erent quality of service guarantees. This provides the rst critical step in detaching the manipulation of the rate from the encoder. An analysis of optimal data partitioning is provided using an operational rate-distortion context, and several algorithms are proposed. The DRS concept then arises by eliminating the second bitstream and allowing the rate constraints to vary over time. DRS provides an interface between the encoder and the network, with which the encoder's output can be perfectly matched to the network's quality of service characteristics. In essence, DRS bridges the gap between constant and variable bit rate video, providing a continuum of possibilities between the two. The problem of optimal DRS is analyzed, and a family of optimal and fast algorithms is described. We also show that Data Partitioning is a special case of clustered constrained" DRS. Some of the fast DRS algorithms perform extremely close to optimal; their low complexity allows even purely software-based real-time implementation, thus making them attractive candidates for incorporation in actual multimedia communication systems.
IEEE Transactions on Circuits and Systems for Video Technology, 2004
We analyze the problem of optimal data partitioning of MPEG-2 coded video in an operational rate-... more We analyze the problem of optimal data partitioning of MPEG-2 coded video in an operational rate-distortion context. The optimal algorithm is characterized and shown to have high complexity and delay. A causally optimal algorithm based on Lagrangian optimization is proposed, that optimally solves the problem for intra (I) pictures, and provides an optimal solution for predicted/interpolated (P/B) pictures when the additional constraints of causal operation and/or low-delay are imposed. A memoryless version of the algorithm, theoretically optimal for intra-pictures only, is shown to perform almost identically but with significantly less computational complexity. Finally, a fast, suboptimal algorithm using purely rate-based optimization is also proposed, and is shown to perform quite close (within 1 dB) to the causally optimal one. Experimental results are provided using actual MPEG-2 coded video bitstreams.
Uploads
Papers by Pankaj Batra