Multiple description coding offers an elegant and competitive solution for data transmission over lossy packetbased networks, with a graceful degradation in quality as losses increase. On the other hand, coding techniques based on redundant transforms give a very promising alternative for the generation of multiple descriptions, mainly due to redundancy inherently given by a transform itself, that offers intrinsic resiliency to losses. In this paper, we show how the partitioning of a generic redundant dictionary can be used to obtain an arbitrary number of multiple complementary, yet correlated descriptions. The most significant terms in the signal representation are drawn from the partitions that better approximate the signal, and distributed into the different descriptions, while the less important ones are alternatively split between the descriptions. As compared to state-of-the-art solutions, such a strategy allows for a better central distortion since atoms in different descriptions are not identical. In the same time, it does not penalize the side distortion significantly, since atoms from the same cluster are likely to be highly correlated. The proposed scheme is applied to the multiple description coding of digital images, and simulation results show increased performances compared to state-of-the-art schemes, both in terms of average distortion, and robustness to loss probability variations.