Next revision
|
Previous revision
Next revision
Both sides next revision
|
research:vslam [2015/05/19 23:29] Prof. Dr. Jörg Stückler created |
research:vslam [2015/05/19 23:42] Prof. Dr. Jörg Stückler |
====== Visual SLAM ====== | ====== Visual SLAM ====== |
| **Contact:** [[members:stueckle]] |
| |
| We pursue direct SLAM techniques that instead of using keypoints, directly operate on image intensities both for tracking and mapping. |
| |
| <html><center></html>{{:research:lsdslam:directvskp.png?500|}}<html></center><br></html> |
| |
| ===== Direct SLAM for Monocular and Stereo Cameras ===== |
| |
| **[[research:lsdslam|LSD-SLAM]]** is a direct SLAM technique for monocular and stereo cameras. The camera is tracked using **direct image alignment**, while geometry is estimated in the form of **semi-dense depth maps**, obtained by **filtering** over many pixelwise stereo comparisons. We then build a **Sim(3) pose-graph of keyframes**, which allows to build scale-drift corrected, large-scale maps including loop-closures. |
| |
| <html><center><iframe width="640" height="360" src="//www.youtube.com/embed/GnuQzP3gty4" frameborder="0" allowfullscreen></iframe></center></html> |
| |
| |
| ===== Direct SLAM for RGB-D Cameras ===== |
| |
| For **[[research:rgb-d_sensors_kinect|SLAM with RGB-D cameras]]** (RGB-D SLAM) we developed a method that also tracks the camera using **direct image alignment**. We optimize a **SE(3) pose-graph of keyframes** to find a globally consistent trajectory and alignment of images. |
| |
| <html> |
| <iframe width="560" height="315" src="//www.youtube-nocookie.com/embed/jNbYcw_dmcQ" frameborder="0" allowfullscreen></iframe> |
| </html> |
| |