Abstract
This paper focuses on the problem of hyper-graph matching,by accoumiting for both unary and higher-order affinity terms.Our method is in line with the linear approx-imate framework while the problem is iteratively solved in discrete space.It is empirically found more efficient than many extant continuous methods.Moreover,it avoids un-known accuracy loss by heuristic mounding step from the continuous approaches.Under weak assumptions,we prove the iterative discrete gradient assignment in general will trap into a degenerating case-an m-circle solution path where m is the order of the problem.A tailored adaptive relaxation mechanism is devised to detect the degenerat-ing case and makes the algorithm converge to a fixed point in discrete space.Evaluations on both synthetic and real-world data corroborate the efficiency of our method.
Error is : local variable 'file_pdfbrieftempa' referenced before assignment