Parallel image transmission and compression Using windowed kriging interpolation

We propose the use of windowed kriging to enable ordinary kriging to interpolate sampled natural images of much higher resolution. Due to its high computational cost, applying ordinary kriging on a whole image can interpolate samples from only 5% of a 64×64 image, while windowed kriging can process images with resolutions higher than 512×512 pixels. Our proposal is motivated by our vision of real-time near-natural collaboration from distributed places based on Distributed Multimedia Plays (DMP) architecture that is realizable in another decade. Windowed kriging allows parallel transmission and compression of sampled natural images to facilitate controlled packets dropping by network nodes for graceful video quality degradation with guaranteed maximum end-to-end delay. Our simple window scheme makes parallel interpolation at the receiver side possible, for which a design of parallel hardware implementation is presented. Typical artifacts due to applying the technique can easily be eliminated completely to produce images of decent quality.