Benchmarking of Blind Video Deblurring Methods on Long Exposure and Resource Poor Settings

Maria Ausilia Napoli Spatafora, Massimo Orazio Spata, Luca Guarnera, Alessandro Ortis, Sebastiano Battiato

Abstract: This paper presents a benchmark evaluation of blind video deblurring methods in specific challenging settings. The employed videos are affected by severe deblurring artifacts and acquisition conditions (e.g., low resolution, high exposure, camera motion, complex scene motion, etc.). An in depth state of the art investigation has been carried out. Then, a specific set of methods based on mathematical optimization with image priors has been involved in our benchmark evaluation. The selected methods have been evaluated quantitatively and qualitatively.