how to get the environment pose of maskfusion

[Zhengjun-Du]

in your paper, you compare your method with MaskFusion. my question is that how to get the the environment pose of maskfusion, since MaskFusion just output the model's pose.