One-to-many WebRTC-based point cloud video conferencing

The developed system supports one-to-many point cloud delivery using a multi-description coding (MDC) approach. In contrast to traditional approaches - which either compress point cloud frames as a whole or apply spatial segmentation in the form of tiles - our solution creates several distinct subsets (descriptions) of sampled points. Each of these descriptions can be encoded separately, resulting in lower processing times due to parallelization. Furthermore, multiple descriptions can be merged together after decoding, resulting in a representation of higher visual quality.