PERSIST: A new probabilistic model for data denoising

Abstract

The paper proposes a more efficient solution for reducing the data noise level, using probability estimation for radius-based spatial inference and sampling techniques.