Since these observed pixels "separate" the unobserved ones from the rest of the graph. ]]>

2. The predicted pixel values will depend on all unobserved pixels, since information will propagate (via the messages) from the observed to the unobserved pixels. ]]>

1. In equation (2), the expression in the product is m_ki (x_i). Is this x_i the same x_i we are maxing over?

2. As far as I understand, the predicted pixel values will only depend on the observed pixels neighboring unobserved pixels. Is that correct?

]]>