A new approach that enables data compression of the support layer while keeping 'structured video' with layer representation is proposed. The proposed video coding scheme consists of interframe prediction and arbitrary shape coding for the prediction error. Each support layer can be updated independently without sending any additional information to the decoder. It enables the decoder to synthesize arbitrarily shaped spatio-temporal images at high compression ratio.
"Data compression for structured video using support layer", Proc. SPIE 2727, Visual Communications and Image Processing '96, (27 February 1996); doi: 10.1117/12.233205; https://doi.org/10.1117/12.233205