Geometry for Computer Vision
data science
about this course
How do computers "see" the world? The mathematical foundation of computer vision lies in
understanding how shapes and images transform, and this course will teach you that
language from the ground up.
Starting with the familiar geometry of high school (Euclidean), we'll discover how different
types of transformations create entirely new geometries. You'll learn how allowing shapes to
be scaled gives us Similarity Geometry, how parallel lines behave in Affine Geometry, and
ultimately how Projective Geometry explains why railroad tracks appear to meet at the
horizon, the same mathematics that lets computers reconstruct 3D scenes from 2D
photographs.
This isn't your typical geometry course. Following Felix Klein's revolutionary approach, we'll
see geometry not as a collection of theorems about triangles and circles, but as the study of
what stays the same when shapes transform. This unified view connects ancient Greek
mathematics to cutting-edge computer vision applications like facial recognition, autonomous
vehicles, and augmented reality.
No advanced mathematics required, just high school algebra, basic calculus, and curiosity
about how mathematical ideas connect. Through concrete examples and visual reasoning,
you'll build intuition for concepts that usually appear only in graduate courses.
This course sets the foundation for Computer Vision II, where you'll apply these geometric
tools to real-world vision problems. Whether you're interested in AI, robotics, computer
graphics, or just want to see mathematics from a fresh perspective,
in these programmes
- computer science
elective · 4-year major, 4-year double major, 3-year major, 3-year double major, 4-year minor, 3-year minor
- data science
elective · 4-year major, 4-year double major, 3-year major, 3-year double major, 4-year minor, 3-year minor, concentration (4-year), concentration (3-year)
planning & teaching
before this
recently taught by
- Shankar SubramanianVisiting Professor, Data Science
Recorded in 2026 T2, 2025 T2. Instructors may change.