Section: New Results

Multiple Birth and Cut Algorithm for Multiple Object Detection

Participant : Guillaume Charpiat.

In collaboration with the Ariana team ( Ahmed Gamal-Eldin, Xavier Descombes and Josiane Zerubia), we developed a new optimization method which we call Multiple Birth and Cut (MBC). It combines the recently proposed Multiple Birth and Death (MBD) algorithm and the Graph-Cut algorithm. MBD and MBC optimization methods are applied to energy minimization of an object based model, the marked point process. The most important advantage of the MBC over MBD is the reduction of number of parameters. By proposing good candidates throughout the selection phase in the birth step, the speed of convergence is increased. In this selection phase, the best candidates are chosen from object sets by a belief propagation algorithm. The algorithm is applied on the flamingo counting problem in a colony [37] , [26] .