MH-BCS-SPL是一种多重假设版本的BCS-SPL，其重建是通过从相邻块中剔除多次预测产生的测量域残差驱动的。(MH-BCS-SPL is a multihypothesis version of BCS-SPL with reconstruction driven by the measurement-domain residual resulting from multiple predictions culled from neighboring blocks.)