JPEG compression metric as a quality-aware image transcoding

Abstract

Transcoding is becoming a preferred technique to tailor multimedia objects for delivery across variable network bandwidth and for storage and display on the destination device. This paper presents techniques to quantify the quality-versus-size tradeoff characteristics for transcoding JPEG images. We analyze the characteristics of images available in typical Web sites and explore how we can perform informed transcoding using the JPEG compression metric. We present the effects of this transcoding on the image storage size and image information quality. We also present ways of predicting the computational cost as well as potential space benefits achieved by the transcoding. These results are useful in any system that uses transcoding to reduce access latencies, increase effective storage space as well as reduce access costs.