Comment on MIMO: Controllable Character Video Synthesis with Spatial Decomposed Modeling