In the course of the lecture "Computer Vision II" by Prof. Leibe at RWTH Aachen University, we discussed contour based tracking as proposed in "Snakes: Active Contour Models" by Kass et al. and used dynamic programming to solve the resulting optimization problem. At this point, Amini et al. give a thorough discussion of how to use dynamic programming to solve variational problems in vision. This can be used to avoid a greedy approach to optimize the energy function of the "snake" (that is, the active contour used for tracking).
What is your opinion on the summarized work? Or do you know related work that is of interest? Let me know your thoughts in the comments below: