Course Contents
Basics of Computer Vision: Introduction to Computer Vision, Geometric Camera Models, Camera Calibration, Extrinsic and Intrinsic Parameters, Essential and Fundamental Matrix, Lighting and Shading - BRDF function (10 Lectures) (Lab: 6 programming hours)
Geometric Vision: Single-view geometry - Projective Transformation in 2D & 3D, Shape from Shading; Two-view geometry - Epipolar Geometry, Stereo Vision; Three-view Geometry - Trifocal tensor (10 Lectures) (Lab: 8 programming hours)
Mid-level Vision Topics: Structure from Motion - Euclidean/Affine/Projective Structure and Motion from two or more images; Structure from Optical Flow; Fitting and Matching - The Hough Transform, Fitting Lines and Curves, RANSAC (14 Lectures) (Lab: 10 programming hours)
Advanced Topics: Multi-layer perceptron, Neural Radiance Field, Gaussian Splatting (8 Lectures)
Learning Outcomes
The students will
-
develop a comprehensive understanding of core principles in computer vision, including camera models, calibration, and geometric transformations.
-
learn about various computer vision algorithms for reconstructing 3D shapes from two or more images by solving inverse problems.
-
get familiar with recent novel view synthesis techniques like Neural Radiance Fields (NeRF) and Gaussian Splatting, and understand how these methods are applied to solve complex computer vision challenges.
Textbooks
-
D. A. Forsyth and J. Ponce. Computer Vision: A Modern Approach (2nd Edition). Prentice Hall, 2011, ISBN-10: 9332550115, ISBN-13: 978-93-325-5011-7 (Indian Edition Available)
-
R. Hartley and A. Zisserman. Multiple View Geometry in Computer Vision. Cambridge University Press, 2002. ISBN-10: 0521540518, ISBN-13: 978-0521540513
Reference Books
-
Berthold Horn. Robot Vision, MIT Press edition, 1986. ISBN-10: 0262081598, ISBN-13:978-0262081597
-
Multiple Papers on MLP, NeRF, Gaussian Splatting
Past Offerings
Course Metadata
| Item | Details |
|---|---|
| Course Title | Geometric Computer Vision |
| Course Code | CS5657 |
| Course Credits | 3-0-2-4 |
| Course Category | PME |
| Proposing Faculty | Avirup Mandal |
| Approved on | Senate of IIT Palakkad |
| Course prerequisites | Linear Algebra and Series (MA1011), Introduction to Programming (CS1020) or equivalent courses |
| Course status | NEW |