Optimal Geometric Model Matching Under Full D Perspective

Matching algorithms use random start local search and a D pose recovery algorithm to nd optimal matches between D object models and D image fea tures An algorithm using only a weak perspective ap proximation to full D perspective solves a subset of the test problems presented A second algorithm al ways uses an iterative D pose algorithm to account for D perspective and solves all test problems includ ing those with varying D perspective A third hybrid algorithm uses weak perspective to direct search and D pose to periodically correct for perspective It is faster than the second A fourth algorithm is a hybrid which also uses a technique called subset convergence to escape from some local optima It performs best on the most di cult matching problems