Publications

Robust and Accurate Line- and/or Point-Based Pose Estimation without Manhattan Assumptions,

Yohann Salaun, Renaud Marlet, and Pascal Monasse, ECCV 2016

Bifocal calibration method using both lines and points to give accurate and robust results both in outdoor and indoor scenes.

Paper Code

Multiscale line segment detector for robust and accurate SfM,

Yohann Salaun, Renaud Marlet, and Pascal Monasse, ICPR 2016

Segment detection method that uses an a contrario multiscale approach for more robust and accurate detections on high resolution pictures

Paper Workshop Code

LSD detection MLSD detection

Robust SfM with Little Image Overlap

Yohann Salaun, Renaud Marlet, and Pascal Monasse, 3DV 2017

Scale estimation method that combines bifocal calibration to obtain the camera global positions. It uses points and lines triplets as well as coplanar constraints that do not need trifocal overlap. It gives accurate results in classical dataset and allow calibration for indoor scenes.

Paper(arXiv) Code