%0 Conference Proceedings %T The Multiscale Line Segment Detector %+ CentraleSupélec %+ imagine [Marne-la-Vallée] %+ Laboratoire d'Informatique Gaspard-Monge (LIGM) %+ École des Ponts ParisTech (ENPC) %A Salaün, Yohann %A Marlet, Renaud %A Monasse, Pascal %< avec comité de lecture %( 1st Workshop on Reproducible Research in Pattern Recognition %B 1st Workshop on Reproducible Research in Pattern Recognition %C Cancun, Mexico %I Springer %P 167 - 178 %8 2016-12-04 %D 2016 %R 10.1007/978-3-319-56414-2_12 %Z Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV]Conference papers %X We propose a multiscale extension of a well-known line segment detector, LSD. We show that its multiscale nature makes it much less susceptible to over-segmentation and more robust to low contrast and less sensitive to noise, while keeping the parameter-less advantage of LSD and still being fast. We also present here a dense gradient filter that disregards regions in which lines are likely to be irrelevant. As it reduces line mismatches, this filter improves the robustness of the application to structure-from-motion. It also yields a faster detection. %G English %2 https://enpc.hal.science/hal-01571615/document %2 https://enpc.hal.science/hal-01571615/file/workshop_mlsd.pdf %L hal-01571615 %U https://enpc.hal.science/hal-01571615 %~ ENPC %~ CNRS %~ LIGM_A3SI %~ PARISTECH %~ LIGM %~ CV_LIGM %~ IMAGINE %~ CVN %~ CENTRALESUPELEC %~ UNIV-PARIS-SACLAY %~ CENTRALESUPELEC-SACLAY %~ ESIEE-PARIS %~ GS-ENGINEERING %~ GS-COMPUTER-SCIENCE %~ UNIV-EIFFEL %~ UPEM-UNIVEIFFEL %~ ESIEE-UNIVEIFFEL %~ TEST3-HALCNRS