000196698 001__ 196698
000196698 005__ 20181203023426.0
000196698 022__ $$a1057-7149
000196698 02470 $$2ISI$$a000329581800015
000196698 0247_ $$2doi$$a10.1109/Tip.2013.2253473
000196698 037__ $$aARTICLE
000196698 245__ $$aFast Geodesic Active Fields for Image Registration Based on Splitting and Augmented Lagrangian Approaches
000196698 269__ $$a2014
000196698 260__ $$bIeee-Inst Electrical Electronics Engineers Inc$$c2014$$aPiscataway
000196698 300__ $$a11
000196698 336__ $$aJournal Articles
000196698 520__ $$aIn this paper, we present an efficient numerical scheme for the recently introduced geodesic active fields (GAF) framework for geometric image registration. This framework considers the registration task as a weighted minimal surface problem. Hence, the data-term and the regularization-term are combined through multiplication in a single, parametrization invariant and geometric cost functional. The multiplicative coupling provides an intrinsic, spatially varying and data-dependent tuning of the regularization strength, and the parametrization invariance allows working with images of nonflat geometry, generally defined on any smoothly parametrizable manifold. The resulting energy-minimizing flow, however, has poor numerical properties. Here, we provide an efficient numerical scheme that uses a splitting approach; data and regularity terms are optimized over two distinct deformation fields that are constrained to be equal via an augmented Lagrangian approach. Our approach is more flexible than standard Gaussian regularization, since one can interpolate freely between isotropic Gaussian and anisotropic TV-like smoothing. In this paper, we compare the geodesic active fields method with the popular Demons method and three more recent state-of-the-art algorithms: NL-optical flow, MRF image registration, and landmark-enhanced large displacement optical flow. Thus, we can show the advantages of the proposed FastGAF method. It compares favorably against Demons, both in terms of registration speed and quality. Over the range of example applications, it also consistently produces results not far from more dedicated state-of-the-art methods, illustrating the flexibility of the proposed framework.
000196698 6531_ $$aAugmented Lagrangian (AL)
000196698 6531_ $$aOperator Splitting
000196698 6531_ $$aNon-convex optimization
000196698 6531_ $$aImage registration
000196698 6531_ $$aGeodesic Active Fields
000196698 6531_ $$aDiffusion equations
000196698 6531_ $$aComputational geometry
000196698 6531_ $$abiomedical image processing
000196698 6531_ $$acomputational geometry
000196698 6531_ $$adiffusion equations
000196698 6531_ $$ageodesic active fields (GAF)
000196698 6531_ $$aimage registration
000196698 6531_ $$anonconvex optimization
000196698 6531_ $$aoperator splitting
000196698 6531_ $$alts
000196698 6531_ $$alts5
000196698 700__ $$0242934$$g148664$$aZosso, Dominique
000196698 700__ $$aBresson, Xavier
000196698 700__ $$aThiran, Jean-Philippe$$g115534$$0240323
000196698 773__ $$j23$$tIeee Transactions On Image Processing$$k2$$q673-683
000196698 909C0 $$xU10954$$0252394$$pLTS5
000196698 909CO $$pSTI$$particle$$ooai:infoscience.tind.io:196698
000196698 917Z8 $$x148664
000196698 937__ $$aEPFL-ARTICLE-196698
000196698 973__ $$rREVIEWED$$sPUBLISHED$$aEPFL
000196698 980__ $$aARTICLE