1. Fast Semi-Direct Monocular Visual Odometry
SVO
Fast Semi-Direct Monocular Visual Odometry
Christian Forster, Matia Pizzoli, Davide Scaramuzza
Shervin Ghasemlou – November 2015

2. Introduction
SVO:
Is a method that combines Feature based methods and direct methods for visual Odometry
According to authors, All Visual Odometry works for MAVs are featurebased.
SVO is more robust and allows faster flight maneuvers
Higher accuracy and speed
Shervin Ghasemlou – December 2015

3. Introduction Feature based Methods: Extract a sparse set of features
Match them in successive frames
Recover camera pose and also structure using epipolar geometry
Finally refine pose and structure
Shervin Ghasemlou – December 2015

4. Introduction Direct Methods: Uses intensity of the image
Exploit information from all parts of the image
These methods outperform feature based mthods in term of robustness
In scenes with little textures
Camera defocus
Motion blur
They save time of feature detection
Shervin Ghasemlou – December 2015

5. Contributions
1-A novel semi direct VO method for MAVs, which in comparison with the state of the art methods, is
faster
More accurate
2-Integration of a probabilistic mapping method
Robust to outliers
Shervin Ghasemlou – December 2015

6. Algorithm Two Thread Motion Estimation Mapping
Sparse model-based image alignment
Feature alignment
Pose and structure refinement
Mapping
Shervin Ghasemlou – December 2015

7. Flowchart
Shervin Ghasemlou – December 2015

8. Motion Estimation (1)Sparse mode based image alignment
The maximum likelihood estimate of the rigid body transformation Tk,k−1 between two consecutive camera poses minimizes the negative log-likelihood of the intensity residuals:
The intensity residual dδ is defined by the photometric difference between pixels observing the same 3D point.
Shervin Ghasemlou – December 2015

9. Motion Estimation
It can be computed by back-projecting a 2D point u from the previous image Ik−1 and subsequently projecting it into the current camera view:
Where
Shervin Ghasemlou – December 2015

10. Motion Estimation
Shervin Ghasemlou – December 2015

11. Motion Estimation (2)Relaxation through feature alignment
The last step aligned the camera with respect to the previous frame
To reduce the drift, the camera pose should be aligned with respect to the map,rather than to the previous frame
Shervin Ghasemlou – December 2015

12. Motion Estimation
Shervin Ghasemlou – December 2015

13. Motion Estimation (3)Pose and Structure Refinement
In this final step, we again optimize the camera pose Tk,w to minimize the re-projection residuals:
Shervin Ghasemlou – December 2015

14. Motion Estimation
Shervin Ghasemlou – December 2015

15. Experimental results
Experiments were performed on a data set recorded from two sources
On the quad-rotor
A handheld camera
Process done on :
Quad-copter embedded platform
A laptop
Two settings
One for accuracy
One for speed
Shervin Ghasemlou – December 2015

16. Experimental results Comparison with PTAM
The proposed method has fewer outliers due to the depth-filter
Shervin Ghasemlou – December 2015

17. Questions?