Date Topic Deliverables Reading
2025-08-25 Introduction and course overview; Reconstruction, Recognition and Synthesis problems
2025-08-27 Reconstruction : Image Formation Geometry and Physics Szeliski - Ch. 2, FoCV - Section II
2025-09-03 Reconstruction : Classical 3D Reconstruction pipelines; Camera Calibration ; Triangulation Szeliski - 11.1, 11.2, FoCV - Section XI, ch. 38-41
2025-09-08 Reconstruction : Epipolar Constraint ; Structure From Motion Szeliski - 11.3 - 11.5, FoCV - Section XI, ch. 44
2025-09-10 Reconstruction : Estimating Correspondences ; RANSAC ; Colmap Szeliski - 11.3 - 11.5, FoCV - Section XI, ch. 44
2025-09-15 Reconstruction : Learning correspondences ; Introducing DUST3R ; DUST3R variants see linked papers from slides
2025-09-17 Reconstruction : Introducing novel view synthesis ; Radiance ; Neural Radiance fields Hw 1 released FoCV - Section XI, ch. 45, see linked papers
2025-09-22 Reconstruction : Gaussian Splats ; Other NeRF variants Project 1 released
2025-09-24 <--Buffer-->
2025-09-29 Recognition : Image classification ; Machine Learning ; Machine learning pipelines ; Neural Networks Szeliski 5.1-5.3, FoCV - Section III
2025-10-01 Recognition : Network architectures : Convolutional Networks ; Transformers Szeliski 5.4 and 5.5, FoCV - Section VII
2025-10-06 Recognition : The data problem : Transfer learning ; Prompt-tuning ; Semi-supervised learning ; Contrastive learning FoCV - Section IX, 30 and Section X
2025-10-08 Recognition : Object Detection : Problem Statement ; R-CNN ; YOLO See linked papers
2025-10-15 Recognition : Semantic Segmentation : Problem Statement ; Architectures and training pipelines ; Dense CRFs and Graph-based segmentation Homework 2 released See linked papers
2025-10-20 Recognition : Structured Prediction ; Pose estimation See linked papers
2025-10-22 Recognition : Open-world recognition ; Vision-language models ; CLIP FoCV - Section XIII, 51, see linked papers
2025-10-27 Recognition : Captioning ; Dense Captioning ; Scene Graphs ; Question-answering See linked papers
2025-10-29 Recognition : Multimodal language models ; Capabilities and limitations ; Question-answering through program generation See linked papers
2025-11-03 <--Buffer--> Project 2 released
2025-11-05 Synthesis : The generative modeling problem ; Technical Challenges FoCV - Section IX, 31, see linked papers
2025-11-10 Synthesis : Neural networks for generation ; GANs FoCV - Section IX, see linked papers
2025-11-12 Synthesis : VAEs FoCV - Section IX, see linked papers
2025-11-17 Synthesis : Diffusion Models FoCV - Section IX, see linked papers
2025-11-19 Synthesis : Conditional generation ; Emergent capabilities ; Novel view synthesis ; Combination with NeRFs FoCV - Section IX, see linked papers
2025-11-24 Videos : Optical flow ; Motion estimation ; Point tracking Szeliski 9, FoCV - Section XII
2025-12-01 Videos : Recognition architectures ; Recognition challenges See linked papers
2025-12-03 Embodied computer vision ; Computer vision with other sensors See linked papers
2025-12-08 Recognition with satellite imagery Take-home exam released See linked papers