Robust compression of 3-D point clouds using multiple description coding

A multiple description coding (MDC) technique for 3-D surface geometry is proposed in this work. The encoder uses a plane-based representation to describe point samples. Then, those plane primitives are classified into two disjoint subsets or two descriptions, each of which provides equal contribution in 3-D surface description. The two descriptions are compressed and transmitted over distinct channels. At the decoder, if both channels are available, the descriptions are decoded and merged together to reconstruct a high quality surface. If only one channel is available, we employ a surface interpolation method to fill visual holes and reconstruct a smooth surface. Therefore, the proposed algorithm can provide an acceptable reconstruction even though one channel is totally lost. Simulation results demonstrate that the proposed algorithm is a promising scheme for 3-D data transmission over noisy channels.