Guest Editorial Special Issue on Advances in Multimedia Computing, Communications and Applications

Abstract

This special issue serves as a forum for recent developments in multimedia signal processing and applications. The ten papers published in this special issue cover a wide range of techniques and applications in various multimedia processing tasks, including multimedia analysis and retrieval, compression and optimization, communication and networking, and multimedia system and applications. In the first paper, Kung and Mak present the paper “PDA-SVM hybrid: a unified model for kernel-based supervised classification.” They have developed a versatile classifier endowed with a broad array of parameters to cope with various kinds of real-world multimedia data. Two papers in the issue explore techniques in media annotation and tagging. Lu and Ip present the relevance model-based framework for an automatic image annotation approach and propose a generalized relevance model to solve the problems with the traditional relevance models. Sandnes then proposes an automatic method for extracting geo-spatial information from the image contents. The elevation of the sun is first estimated indirectly by measuring the relative length of objects and their shadows in image scenes, and then combined with the creation time of the image for estimating the approximate geographical location of the photographer. The next three papers investigate visual processing techniques in multimedia research for object detection and tracking as well as for face recognition. Huang and Li present two new techniques: color-salience weighted probability-of-boundary (cPoB) and successive classification and refinement (SCR) for automatic object detection and tracking in active video. Khan et al. propose a robust visual object tracking scheme that uses the dynamics of object shape and appearance similarity. The method uses a particle filter combined with a multi-mode anisotropic mean shift to perform tracking. Lu et al. investigate the face recognition problem by using a 2-D linear discrimination analysis approach. In this work, they contribute to the optimization of the weighted matrix in the discriminator and also give an iterative algorithm for evaluating the weighting matrix. Another group of three papers concentrate on multimedia compression and optimization. Lin and Hsu present intraframe transcoding architecture for MPEG-2/H.263/H.264/ MPEG-4 AVC to H.264/MPEG-4 AVC via discrete cosine transform coefficients and coding modes. An I4MB/I16MB mode selection is employed by the summation of the low frequencyAC coefficients. An I4MBmode refinement scheme is designed to eliminate the unlikely intra prediction modes for H.263 and H.264/MPEG-4 AVC coding stream. Wang et al. introduce a macro-block group (MBG) parallelism for multicore processors in the AVS coding scheme and compare the coding performance with that of slice-level parallelism. Yeh et al. describe a channel selection algorithm to enhance the P. Muneesawang (*) Naresuan University, Phisanulok 65000, Thailand e-mail: paisarnmu@nu.ac.th