Where to start
Course overview
Understand the course structure, schedule, and grading.
Bibliography
Key textbooks and reference materials used throughout the course.
Computational geometry
Homogeneous coordinates, transformations, calibration, and multi-view geometry.
Deep learning
CNNs, YOLO, facial analysis, segmentation, GANs, and Transformers.
Course modules
Homogeneous coordinates
Points, lines, and planes in projective space.
2D & 3D transformations
Euclidean, similarity, affine, and projective transformations.
Camera calibration
Intrinsic/extrinsic parameters, RANSAC, and least-squares estimation.
Epipolar & trifocal geometry
Fundamental matrix, 3D reconstruction, and trifocal tensors.
CNNs & image classification
Convolutional neural networks and transfer learning with PyTorch.
Object detection — YOLO
Real-time detection, tracking, and anomaly detection.
GANs & Stable Diffusion
Generative adversarial networks and diffusion-based image synthesis.
Ethics & fairness
Bias, explainability, adversarial attacks, and responsible AI.
Getting started
Review the course overview
Read the course overview to understand the schedule, topics, and expectations.
Set up your Python environment
Most hands-on exercises run in Google Colab — no local setup required. Open any Colab link directly from the topic pages.
Start with geometry
Begin with homogeneous coordinates to build the mathematical foundation needed for the rest of the course.
Explore deep learning
Once comfortable with geometry, move to deep learning to work with CNNs, YOLO, and Transformers.
Recorded lectures from the 2021 edition of the course are linked throughout the materials as supplementary viewing. Look for Video: Clase grabada references on each topic page.