Epipolarna - Project Presentation - 3D Reconstruction
-
Upload
mattiastiger -
Category
Engineering
-
view
46 -
download
5
Transcript of Epipolarna - Project Presentation - 3D Reconstruction
![Page 1: Epipolarna - Project Presentation - 3D Reconstruction](https://reader035.fdocuments.in/reader035/viewer/2022062420/55c0e98bbb61eb77678b46d5/html5/thumbnails/1.jpg)
TSBB15 3D Reconstruction Group 2 “Epipolarna”
![Page 2: Epipolarna - Project Presentation - 3D Reconstruction](https://reader035.fdocuments.in/reader035/viewer/2022062420/55c0e98bbb61eb77678b46d5/html5/thumbnails/2.jpg)
Modules
Main program and pipeline – Mattias
2D correspondence extraction – Alexander
Non-linear optimization – Martin
Visualization – Gustav
![Page 3: Epipolarna - Project Presentation - 3D Reconstruction](https://reader035.fdocuments.in/reader035/viewer/2022062420/55c0e98bbb61eb77678b46d5/html5/thumbnails/3.jpg)
The 3D Reconstruction pipeline
![Page 4: Epipolarna - Project Presentation - 3D Reconstruction](https://reader035.fdocuments.in/reader035/viewer/2022062420/55c0e98bbb61eb77678b46d5/html5/thumbnails/4.jpg)
Data Structures
Camera data structure
2D <-> 3D
Camera pair & 2D <-> 3D
![Page 5: Epipolarna - Project Presentation - 3D Reconstruction](https://reader035.fdocuments.in/reader035/viewer/2022062420/55c0e98bbb61eb77678b46d5/html5/thumbnails/5.jpg)
Problems
Difficult data set
Outliers
Long time before results are seen
![Page 6: Epipolarna - Project Presentation - 3D Reconstruction](https://reader035.fdocuments.in/reader035/viewer/2022062420/55c0e98bbb61eb77678b46d5/html5/thumbnails/6.jpg)
Improvements
Chains?
1 2 3
![Page 7: Epipolarna - Project Presentation - 3D Reconstruction](https://reader035.fdocuments.in/reader035/viewer/2022062420/55c0e98bbb61eb77678b46d5/html5/thumbnails/7.jpg)
2D point extraction
Find feature points, using Harris response Minimum relative quality Minimum distance
Calculate descriptor, using SIFT
Calculate correspondences, using Brute Force
![Page 8: Epipolarna - Project Presentation - 3D Reconstruction](https://reader035.fdocuments.in/reader035/viewer/2022062420/55c0e98bbb61eb77678b46d5/html5/thumbnails/8.jpg)
Feature Points
![Page 9: Epipolarna - Project Presentation - 3D Reconstruction](https://reader035.fdocuments.in/reader035/viewer/2022062420/55c0e98bbb61eb77678b46d5/html5/thumbnails/9.jpg)
Corresponding Points
![Page 10: Epipolarna - Project Presentation - 3D Reconstruction](https://reader035.fdocuments.in/reader035/viewer/2022062420/55c0e98bbb61eb77678b46d5/html5/thumbnails/10.jpg)
Initial Pose
Initial relation between cameras is calculated from the essential matrix, according to algorithm described in an epic compendium by Klas Nordberg.
Results used as initial guess for PnP
![Page 11: Epipolarna - Project Presentation - 3D Reconstruction](https://reader035.fdocuments.in/reader035/viewer/2022062420/55c0e98bbb61eb77678b46d5/html5/thumbnails/11.jpg)
Non-linear optimization
Implemented using the levmar API
Called three times in one pipeline iteration Last step of Gold Standard algorithm PnP pose estimation Bundle adjustment
Minimizes squared re-projection error over specified parameters.
![Page 12: Epipolarna - Project Presentation - 3D Reconstruction](https://reader035.fdocuments.in/reader035/viewer/2022062420/55c0e98bbb61eb77678b46d5/html5/thumbnails/12.jpg)
Rotation parameterization
Vector representation
Support exists in OpenCV: (cv::Rodrigues)
Allows for unconstrained optimization
Ambiguity in the representation
![Page 13: Epipolarna - Project Presentation - 3D Reconstruction](https://reader035.fdocuments.in/reader035/viewer/2022062420/55c0e98bbb61eb77678b46d5/html5/thumbnails/13.jpg)
Solving the PnP
Initial guess derived from F (R & t)
Use known correspondences from previous view
Threshold new points on re-projection error to remove outliers
![Page 14: Epipolarna - Project Presentation - 3D Reconstruction](https://reader035.fdocuments.in/reader035/viewer/2022062420/55c0e98bbb61eb77678b46d5/html5/thumbnails/14.jpg)
Bundle Adjustment
Same rotation parameterization as the one used for solving the PnP
Computation time scales badly with the number of points and views Dinosaur set takes more than ten hours to complete
(30 points per view)
Slow mainly because No sparse pattern is used for Jacobian estimation levmar was not built using LAPACK
![Page 15: Epipolarna - Project Presentation - 3D Reconstruction](https://reader035.fdocuments.in/reader035/viewer/2022062420/55c0e98bbb61eb77678b46d5/html5/thumbnails/15.jpg)
Visualization
Plot the estimated cameras
Plot the 3D points.
Points colored from original image data.
![Page 16: Epipolarna - Project Presentation - 3D Reconstruction](https://reader035.fdocuments.in/reader035/viewer/2022062420/55c0e98bbb61eb77678b46d5/html5/thumbnails/16.jpg)
Results
Very slow for many cameras.Some drift for the last few cameras.Good results except for the last 3 cameras.