A variational approach to multi-modal image matching

摘要: We address the problem of nonparametric multi-modal image matching. propose a generic framework which relies on global variational formulation and show its versatility through three different registration methods: supervised by joint intensity learning, maximization mutual information correlation ratio. Regularization is performed using functional borrowed from linear elasticity theory. also consider geometry-driven regularization method. Experiments synthetic images preliminary results realignment MRI datasets are presented.