A fast scheme for arbitrarily resizing of digital image in the compressed domain

Arbitrarily resizing of digital images is needed in many applications, such as transmission of images over communication channels whose bandwidth is varying widely, or display of the pre-encoding program on devices with different resolution. This paper presents a new algorithm, which possesses the capability of arbitrarily image resizing in compressed domain. Experimental results confirm that this scheme can achieve significant computation cost reduction while maintaining better quality. Hence, it can be efficiently used in standalone image resizing and in transcoding applications which require real-time processing.