Alexandru Telea

TL;DR: A new algorithm for digital inpainting based on the fast marching method for level set applications is presented, very simple to implement, fast, and produces nearly identical results to more complex, and usually slower, known methods.

TL;DR: It is shown how visualization can provide highly valuable feedback for network designers through experiments conducted in three traditional image classification benchmark datasets, and the presence of interpretable clusters of learned representations and the partitioning of artificial neurons into groups with apparently related discriminative roles are discovered.

TL;DR: A taxonomy of 3D skeletons is proposed which allows us to further analyze and compare them with respect to their properties, and describes several applications of3D skeletons, which illustrate their added value for different shape analysis and processing tasks.

TL;DR: This work characterize the input data space, projection techniques, and the quality of projections, by several quantitative metrics, and samples these three spaces according to these metrics, aiming at good coverage with bounded effort.

TL;DR: A novel approach that combines edge clustering, distance fields, and 2D skeletonization to construct progressively bundled layouts for general graphs by iteratively attracting edges towards the centerlines of level sets of their distance fields.