Abstract

This paper advocates a novel material-aware feature descrip- tor for volumetric image registration. We rigorously formulate a novel probability density function (PDF) based distance metric to devise a compact local feature descriptor supporting invariance of full 3D orienta- tion and isometric deformation. The central idea is to employ anisotropic heat diffusion to characterize the detected local volumetric features. It is achieved by the elegant unification of diffusion tensor (DT) space con- struction based on local Hessian eigen-system, multi-scale feature extrac- tion based on DT-weighted dyadic wavelet transform, and local distance definition based on PDF formulated in DT space. The diffusion, intrin- sic structure-aware nature makes our volumetric feature descriptor more robust to noise. With volumetric images registration as verifiable appli- cation, various experiments on different volumetric images demonstrate the superiority of our descriptor.