Computer Vision

Lectures: on Mondays, 8:25-9:55 , lecture hall 0.79  0.79 Jánossy Lajos lecture hall,  (North Building)

Practices:

• On Wednesdays 17.55-19.25, South building at
  • South building 0-311 Konig terem room
  • South building 0-312 Gallai Tibor room
  • (South building 0-220 Kárteszi Ferenc room)

Teachers:

• Levente Hajder
• Tamás Tófalvi
• Tarlan Ahadli

Demonstrators

• Máté Poór
• Muhammad Rafi Faisal

Agenda

Week	Lecture	Laboratory
1st – Registration week
2nd	Introduction	Vehicle and sensor kit demonstration ( videos of 2023)
3rd	Estimation theory: inhomogeneous linear systems	OpenCV installation, GUI by OpenCV
4th	Lagrange multipliers; homogeneous linear systems, point-line/plane distances	Affine Transformations
5th	Random Sampling Consensus (RANSAC)	RANSAC (Whiteboard)
6th	Optimal Plane Fitting; SVD; Camera models	Pointset visualization (Whiteboard)
7th	Back-projection. Introduction to homographies.	Multi-model fitting
8th	Homography Estimation. Data Normalization.	National holiday (break)
9th – fall break
10th	Camera Calibration, RQ-decomposition.	Homography Estimation
11th	Introduction to Stereo Vision	Point registration: theory & source code
12th	Stereo vision (cont)	Tomasi-Kanade factorization (theory, source: Win, Linux)
13th	Stereo vision (cont)	Camera Calibration
14th	Planar motion Numerical optimization	Stereo pipeline
15th	Bundle adjustment. Special hardwares.	(sources in Canvas)

Oral exams:

• 18th December 11:00
• 3rd January 10:00
• 7th January 10:00
• 17th January 10:00
• 24th January 10:00
• 31th January 16:00

Assignment presentations:

• 16th December 16:00 (online)
• 2nd January 16:00 (online)
• 6th January 16:00 (online)
• 16th January 16:00 (online)
• 23rd January 16:00 (online)
• 31th January 16:00 (in person, during the oral exam)

Topics for oral exams

1. Basic estimation theory: solution for homogeneous and inhomogeneous linear system of equations.
2. Robust fitting methods: the RANSAC method.
3. Multi-model fitting: sequential RANSAC, MultiRANSAC
4. Camera models: perspective camera, orthogonal projection, weak-perspective camera.
5. Calibration of perspective camera using a spatial calibration object: Estimation and decomposition of projection matrix.
6. Chessboard-based calibration of perspective cameras. Radial and tangential distortion.
7. Plane-plane homography. Panoramic images by homographies.
8. Estimation of homography. Data normalization.
9. Basics of epipolar geometry. Essential and fundamental matrices. Rectification.
10. Estimation of essential and fundamental matrices. Data normalization. Decomposition of essential matrix.
11. Triangulation for both standard and general stereo vision.
12. Stereo vision for planar motion.
13. Tomasi-Kanade factorization: multi-view reconstruction by orthogonal and weak-perspective camera models.
14. Reconstruction by merging stereo reconstructions. Registration of two point sets by a similarity transformation.
15. Numerical optimization
16. Numerical multi-view reconstruction: bundle adjustment.
17. Tomasi-Kanade factorization with missing data.
18. Reconstruction by special devices: laser scanning, depth camera, LiDAR.

Final grade

The final grade is the sum of oral exam (max. 100%) and assignment scores. At least 40% should be reached both from oral exam and assignment scrores.

Thresholds for marks as follows:

• Excellent (5): >=170%
• Good (4): 140-169%
• Satisfactory (3): 110-139%
• Pass (2): 80-109%
• Fail (1): 0-79%