50:49Joan Lasenby on Applications of Geometric Algebra in Engineering
From Y Combinator · Published Jul 22, 2019 · Watch on YouTube
TL;DR
Geometric algebra (GA) unifies scalars, vectors, bivectors, and trivectors into a single algebraic system that simplifies rotations (via rotors), coordinate-free expressions, and differentiation of geometric objects. It was historically developed by Grassmann, Clifford, and David Hestenes, and includes conformal GA (5D) where points, lines, circles, and spheres are first-class objects.
Key insights
- Lines are harder than points in classical computer vision; GA makes line-based reconstruction natural because lines are objects in the algebra.
- Cross product (a × b) works only in 3D; GA’s outer product (wedge) generalizes to any dimension via bivectors, trivectors, etc.
- Quaternions (i, j, k) are actually unit bivectors (oriented planes) in 3D, not imaginary numbers; GA explains them as rotors.
Want the full analysis - every claim cited to the second it was said?
This page only shows a teaser. Sign up to chat with the complete, cited breakdown of "Joan Lasenby on Applications of Geometric Algebra in Engineering".