Abstract. We use a simple yet powerful higher-order conditional random field (CRF) to model optical flow. It consists of a standard photoconsistency cost and a prior on affine motions both modeled in terms of higher-order potential functions. Reasoning jointly over a large set of unknown variables provides more reliable motion estimates and a robust matching criterion. One of the main contributions is that unlike previous region-based methods, we omit the assumption of constant flow. Instead, we consider local affine warps whose likelihood energy can be computed exactly without approximations. This results in a tractable, so-called, higher-order likelihood function. We realize this idea by employing triangulation meshes which immensely reduce the complexity of the problem. Optimization is performed by hierarchical QPBO moves and an adaptive mesh refinement strategy. Experiments show that we achieve high-quality motion fields on several data sets including the Middlebury optical flow database. 1

We propose a Markov Random Field formulation for the tracking of needles in fluoroscopic images. A novel motion model makes it possible to capture the primarily rigid motion as well as deformations of the needle in a single second-order MRF graph. Needles are represented by B-splines and each control point is associated with a random variable in a MAP-MRF formulation. In addition to the control points we introduce a single additional random variable representing the rigid transformation needles undergo during interventions. The incorporation of rigid transformations allows to recover transformations even in the presence of large displacements which is not possible with existing MRF models for medical tool tracking.

c t i v it y e p o r t 2009 Table of contents 1. Team.................................................................................... 1

apport de recherche