User Tools

Site Tools


people:kidson:visual_odometry

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Next revision
Previous revision
people:kidson:visual_odometry [2013/02/12 22:51]
kidson created
people:kidson:visual_odometry [2013/05/30 21:14] (current)
kidson
Line 8: Line 8:
  
 The bottom video shows the estimated position of the car (estimate marker) together with the ground truth position (kitty_stereo_left marker). The bottom video shows the estimated position of the car (estimate marker) together with the ground truth position (kitty_stereo_left marker).
 +
 +The white dots are visual features extracted from the scene.
  
 <​html>​ <​html>​
Line 23: Line 25:
 </​html>​ </​html>​
  
-==Links==+The above videos are both running the same SLAM pipeline. ​ 2D features are tracked using a combination of FAST/FREAK and matched efficiently by doing a restricted search (assuming features do not move far).  Features are projected into 3D space using either 3D kinect data or 3D coordinates as calculated from stereo disparity. ​ RANSAC is used to filter outliers and calculate a transformation of the camera between frames. ​ No global optimization has been incorperated into this framework.
  
-[[https://​github.com/​tum-uav/​rvc_waitinglist/​tree/​ross/​|Link to the code]]+==Links==
  
 +[[https://​github.com/​rosskidson/​visual_odometry|Link to the code]]
  
  
people/kidson/visual_odometry.1360705906.txt.gz ยท Last modified: 2013/02/12 22:51 by kidson