Direct Stereo Visual Odometry Based on Lines

Thomas Holzmann; Friedrich Fraundorfer; Horst Bischof

Direct Stereo Visual Odometry Based on Lines

Thomas Holzmann, Friedrich Fraundorfer, Horst Bischof

Institute of Computer Graphics and Vision (7100)

Research output: Chapter in Book/Report/Conference proceeding › Conference paper › peer-review

Abstract

We propose a novel stereo visual odometry approach, which is especially suited for poorly textured environments. We introduce a novel, fast line segment detector and matcher, which detects vertical lines supported by an IMU. The patches around lines are then used to directly estimate the pose of consecutive cameras by minimizing the photometric error. Our algorithm outperforms state-of-the-art approaches in challenging environments. Our implementation runs in real-time and is therefore well suited for various robotics and augmented reality applications.

Original language	English
Title of host publication	11th International Conference on Computer Vision Theory and Applications (VISAPP), 2016
Publication status	Published - Feb 2016
Event	11th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications: VISIGRAPP 2016 - Rome, Italy Duration: 27 Feb 2016 → 29 Feb 2016

Conference

Conference	11th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications
Country/Territory	Italy
City	Rome
Period	27/02/16 → 29/02/16

Access to Document

line_vo

Cite this

@inproceedings{817443d4a0bf4e49b814683eead04929,

title = "Direct Stereo Visual Odometry Based on Lines",

abstract = "We propose a novel stereo visual odometry approach, which is especially suited for poorly textured environments. We introduce a novel, fast line segment detector and matcher, which detects vertical lines supported by an IMU. The patches around lines are then used to directly estimate the pose of consecutive cameras by minimizing the photometric error. Our algorithm outperforms state-of-the-art approaches in challenging environments. Our implementation runs in real-time and is therefore well suited for various robotics and augmented reality applications.",

author = "Thomas Holzmann and Friedrich Fraundorfer and Horst Bischof",

year = "2016",

month = feb,

language = "English",

booktitle = "11th International Conference on Computer Vision Theory and Applications (VISAPP), 2016",

note = "11th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications : VISIGRAPP 2016 ; Conference date: 27-02-2016 Through 29-02-2016",

}

TY - GEN

T1 - Direct Stereo Visual Odometry Based on Lines

AU - Holzmann, Thomas

AU - Fraundorfer, Friedrich

AU - Bischof, Horst

PY - 2016/2

Y1 - 2016/2

N2 - We propose a novel stereo visual odometry approach, which is especially suited for poorly textured environments. We introduce a novel, fast line segment detector and matcher, which detects vertical lines supported by an IMU. The patches around lines are then used to directly estimate the pose of consecutive cameras by minimizing the photometric error. Our algorithm outperforms state-of-the-art approaches in challenging environments. Our implementation runs in real-time and is therefore well suited for various robotics and augmented reality applications.

AB - We propose a novel stereo visual odometry approach, which is especially suited for poorly textured environments. We introduce a novel, fast line segment detector and matcher, which detects vertical lines supported by an IMU. The patches around lines are then used to directly estimate the pose of consecutive cameras by minimizing the photometric error. Our algorithm outperforms state-of-the-art approaches in challenging environments. Our implementation runs in real-time and is therefore well suited for various robotics and augmented reality applications.

M3 - Conference paper

BT - 11th International Conference on Computer Vision Theory and Applications (VISAPP), 2016

T2 - 11th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications

Y2 - 27 February 2016 through 29 February 2016

ER -