Lectures / Notes:

Below is the (tentative) list of classes, with possible additional readings. These may change as the semester progresses.
Date Topic Deliverables Readings Additional papers/reading
8/29/19 Introduction
9/3/19 Image formation - geometry
[pptx|pdf]
Szeliski 2.1
9/5/19 Image formation - color
[pptx | pdf]
Szeliski 2.2
9/10/19 Basic image processing
[pptx | pdf]
Szeliski 3.1-3.5, 4.2-4.3, SIRFS
9/12/19 Grouping Normalized cuts Random forest-based
Neural network-based
9/17/19 Reconstruction 1
9/19/19 Reconstruction 2 Project proposal due A taxonomy and evaluation of dense two-frame stereo correspondence algorithms
9/24/19 Reconstruction 3
9/26/19 Correspondence SIFT UCN
WarpNet
10/1/19 Optical Flow High accuracy optical flow estimation based on a theory of warping
10/3/19 Intro to ML | Neural Networks 1 LeCun '98 (Till Section IV)
10/8/19 Neural networks 2 | Transfer learning Do better imagenet models transfer better?
Residual adapters
10/10/19 Neural networks 3 | Learning under resource constraints Project check-in (updated project proposal) Residual networks MSDNets
Mobile Net v2
10/15/19 No class - Fall break
10/17/19 Semantic segmentation U-net Deeplab v3
10/22/19 Structured prediction
10/24/19 Graphical models for refinement Autocontext Inference machines
CRF as RNN
10/29/19 Object detection
10/31/19 Instance segmentation Mask R-CNN
11/5/19 Pose estimation Convolutional Pose estimation, Associative embedding
11/7/19 Learning 3D PTN
11/12/19 Vision and Language
11/14/19 Reinforcement learning Target-driven Visual Navigation in Indoor Scenes using Deep Reinforcement Learning
11/19/19 Embodied cognition Preliminary project report
11/21/19 Generative modeling
11/26/19 Learning without supervision Preliminary project reviews
11/28/19 No class - Thanksgiving
12/3/19 Bias and ethics student presentations Women also snowboard
12/5/19 Student presentations / Q&A
12/10/19 student presentations / Q&A
12/19/19 final report, assignment due