Interactive 2D/3D Image Denoising and Segmentation Tool for Medical Applications

We present an interactive image and volume denoising and segmentation tool, that allows the use of state of the art algorithms based on the concept of total variation (TV) to be applied to 2D and 3D grey and color value data. Our aim is to provide a research tool for experimenting with TV related methods to the medical image analysis community. Our implementation supports the classic ROF and TVL1 denoising models, second order TV denoising variants as well as an interactive segmentation based on weighted TV. The numerical optimization procedures to solve the variational problems follow the primal-dual scheme of Chambolle and Pock. A highly efficient GPU implementation based on NVidia CUDA enables real-time feedback even for large volumetric data sets. We will provide a binary software tool to the public domain for interested researchers at our website and intend to perform a live demonstration on a notebook at the MICCAI workshop.