“…The under-determined OF constraint equation is solved by variational principles in which some other regularization constraints are added in, including the image gradient, the phase or block matching. Although efforts have been made to seek more accurate regularization terms, OF approaches lack accuracy, especially for t-MRI motion tracking, due to the tag fading and large deformation problems [11,49]. More recently, convolutional neural networks (CNN) are trained to predict OF [16,19,20,24,26,41,31,47,53,51,48].…”