Multiple description (MD) coding allows for the redundant coding of multimedia streams while exploiting (virtual) path diversity. This chapter discusses the most popular enabling technologies such as MD scalar quantization, the usage of a correlating transform, the polyphase transform with selective quantization and MD forward error correction. Furthermore, the concept of scalable MD coding is introduced and illustrated in a scalable video coder, based on motion compensated temporal filtering and that incorporates MD technology for the encoding of both texture and motion vector data. The principles of channel-aware MD rate-allocation are detailed as well. Finally, experimental results prove that MD coding can be beneficial for video transmission systems that suffer from poor channel quality and that allow to the exploitation of the path diversity principle.