Multiscale line segment detector for robust and accurate SfM

Salaün, Yohann; Marlet, Renaud; Monasse, Pascal

doi:10.1109/icpr.2016.7899930

Cited by 23 publications

(16 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For all experiments, line segments are detected with a variant of LSD [28] that detects lines at different scales and reduces over-segmentation [29]. They are then matched with LBD [30].…”

Section: Methodsmentioning

confidence: 99%

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

Salaün

Marlet

Monasse

2016

Computer Vision – ECCV 2016

Self Cite

View full text Add to dashboard Cite

Abstract. Usual Structure from Motion techniques based on feature points have a hard time on scenes with little texture or presenting a single plane, as in indoor environments. Line segments are more robust features in this case. We propose a novel geometrical criterion for two-view pose estimation using lines, that does not assume a Manhattan world. We also define a parameterless (a contrario) RANSAC-like method to discard calibration outliers and provide more robust pose estimations, possibly using points as well when available. Last, we provide quantitative experimental data that illustrate failure cases of other methods and that show how our approach outperforms them, both in robustness and accuracy.

show abstract

“…For all experiments, line segments are detected with a variant of LSD [28] that detects lines at different scales and reduces over-segmentation [29]. They are then matched with LBD [30].…”

Section: Methodsmentioning

confidence: 99%

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

Salaün

Marlet

Monasse

2016

Computer Vision – ECCV 2016

Self Cite

View full text Add to dashboard Cite

show abstract

“…We use the same notations as the companion paper [3], recalled here. In the following, the k index will denote the scale associated with the feature.…”

Section: Notationmentioning

confidence: 99%

“…For this, we use a multiscale framework and information from coarser scales to better detect segments at finer scales. In our companion paper [3], we compare it to other state-of-the-art line detectors, namely LSD [2] and EDLines [1], as a building block of a structure from motion (SfM) pipeline [4] to obtain quantitative, objective results.…”

Section: Introductionmentioning

confidence: 99%

The Multiscale Line Segment Detector

Salaün

Marlet

Monasse

2017

Reproducible Research in Pattern Recognition

Self Cite

View full text Add to dashboard Cite

Abstract. 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.

show abstract

“…Lines with longer lengths are prone to over-segmentation in higher resolutions. The method proposed in [52] that is called MLSD (Multiscale LSD), is a multiscale approach based on LSD.…”

Section: Line Detectionmentioning

confidence: 99%

Analysis of Camera Pose Estimation Using 2D Scene Features for Augmented Reality Applications

Alsadat

Laurendeau

2018

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Maîtrise en génie électriqueMaître ès sciences (M. model, the camera has to be localized regarding its surroundings. Camera localization consists of finding the exterior parameters (i.e. its position and orientation) of the camera with respect to the viewed scene and its characteristics.ScIn this thesis, camera pose estimation methods using circle-ellipse and straight line correspondences has been investigated. Circles and lines are two of the geometrical features that are mostly present in structures and buildings. Based on the relationship between the 3D features and their corresponding 2D data detected in the image, the position and orientation of the camera is estimated.

show abstract

Multiscale line segment detector for robust and accurate SfM

Cited by 23 publications

References 13 publications

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

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

The Multiscale Line Segment Detector

Analysis of Camera Pose Estimation Using 2D Scene Features for Augmented Reality Applications

Contact Info

Product

Resources

About