Permuted smoothed descriptions and refinement coding for images

We consider the problem of transmitting compressed still images over lossy channels. In particular, we examine the situation where the data stream is partitioned into two independent channels, as is often considered in the mu ltiple descriptions approach to image compression. In this paper, we introd uce a coder design called smoothed descriptions, which matches a data parti tioning method utilized at the encoder to the error concealment technique e mployed at the decoder. This approach has the advantage of inserting minima l overhead into the transmitted data streams, so that system performance is undiminished when there are no packet losses over the channel, We show tha t, by using a combination of DC averaging and maximal smoothing to conceal errors, performance comparable to or better than multiple descriptions can be achieved for packet loss rates up to 5%. By adding a feedback loop that requests retransmission of important data, we also demonstrate the need to exploit latency whenever possible.