Computer Vision
Lectures: on Tuesdays, 10:15-11:45 , lecture hall -1.62 (North Building)
Practices:
- On Wednesdays 16.15-17.45,
- North building, room 0.99
Teachers:
- Peter Kozma
- Levente Hajder
- Iván Eichhardt
- Tarlan Ahadli
Agenda
| Week | Tuesday |
| 1st: Registration week | |
| 2nd | Introduction |
| 3rd | Estimation theory: inhomogeneous linear systems, homogeneous sytems, circle estimation |
| 4th | LiDAR-Camera calibration using a spherical target |
| 5th | Point set registration |
| 6th | ByAhadliTartlan |
| 7th | ByAhadliTartlan |
| 8th | LiDAR-Camera MATLAB toolbox using chessboards |
| 9th – fall break | |
| 10th | ByAhadliTartlan |
| 11th | ByAhadliTartlan |
| 12th | Lie algebra |
| 13th | Lie algebra (cont.) |
| 14th | Point-ellipse distance. Iterative & Weighted Least-Squares Estimation |
Peter Kozma’s lecture slides:
- Lecture #1: Self-driving Future
- Lecture #2: Camera Sensors
- Lecture #3: LiDAR Sensors
- Lecture #4: Radar Sensors
- Lecture #5: GNSS-INS Sensors
Exams
- 18th December 11:00
- 3rd January 16:00 (room 0-409, South Building)
- 17th January 16:00 (room 2-502, South Building)
- 31st January 16:00 (room 0-412, South Building)
Topics for oral exam (preliminary)
At the oral exam, two topics will be randomly selected.
Qestions from Peter Kozma:
1. Self-driving future.
2. Camera.
3. LiDAR.
4. RADAR.
5. GNSS-INS.
Questions from Iván Eichhardt/Levente Hajder
- Over-determined linear system of equations. Circle and ellipse fitting.
- Application of the L1 norm; the iterative reweighted least squares algorithm.
- Optimal point registration algorithm.
- Sphere radius estimation on LiDAR and camera. Pose (rotation+translation) estimation between a camera and a LiDAR device.
- Lie theory – SO(3): Describe the relation of skew symmetric matrices, infinitesimal rotation. How to derive the exponential map. Closed from for the exponential and logarithmic maps. Alternative representations of 3D rotation.
- Lie theory – SE(3): Describe & define notable linear groups (GL(n) etc.). Representation of rigid motion, Exp, Log. Application(s) of SE(3).
- Lie theory – Applications: Describe the relation of a Lie group and its corresponding Lie algebra. Why is the Identity element of the group so important? Exp, Log, boxplus, boxminus. Interpolation. Uncertain 3D rotation, composition…
Final grade
The final grade is the sum of oral exam (max. 25+25=50) and assignment scores. At least 20 points should be reached both from oral exam and assignment scores.
Thresholds for marks as follows:
- Excellent (5): >=85
- Good (4): 70-84
- Satisfactory (3): 55-69
- Pass (2): 40-54
- Fail (1): 0-39