Document Detection in Videos Captured by Smartphones using a Saliency-Based Method

Minh On Vu Ngoc, Jonathan Fabrizio, Thierry Géraud

Abstract: Smartphones are now widely used to digitize paper documents. Document detection is the first important step of the digitization process. Whereas many methods extract lines from contours as candidates for the document boundary, we present in this paper a region-based approach. A key feature of our method is that it relies on visual saliency, using a recent distance existing in mathematical morphology. We show that the performance of our method is competitive with state-ofthe- art methods on the ICDAR Smartdoc 2015 Competition dataset.