Announcements
01/04/2022: Welcome to the course!01/04/2022: Homework 0 is released on Piazza, due 01/12/2022 23:59 PM
General Information
Times & PlacesTuTh 3:30PM - 4:50PM, Zoom link on Piazza
Course Staff
| Name | Office Hours | Location | ||
|---|---|---|---|---|
| Instructor | Hao Su | haosu@eng.ucsd.edu | Tuesday 2-3 PM | see Piazza |
| Teaching Assistant | Jiayuan Gu | jigu@eng.ucsd.edu | Monday 2-3PM | see Piazza |
| Teaching Assistant | Xiaoshuai Zhang | xiz040@eng.ucsd.edu | Thursday 1PM-2PM | see Piazza |
Objectives
This is a graduate level course to teach state-of-the-art concepts and algorithms of geometry that are being used in computer graphics, computer vision and machine learning. It should enable you to read and replicate recent 3D papers in top CV/CG conferences (not industry job oriented).Prerequisites
- Skilled in linear algebra
- Familiar with Multi-Variable Calculus
- Familiar with Probability and Numerical Methods
- Strong programming skills (Linux toolchain, Python, Numpy, PyTorch)
Grading
- Homework 0 5%
- Homework 1 20%
- Homework 2 20%
- Homework 3 20%
- Final project 35%
- Extra credit for participation 5% (ask/answer questions in class, attend office hours)
- There will not be a final exam.
Syllabus
- Geometry Basics
- 1D/2D/3D Geometry
- Transformation
- Storing Geometry in Computer
- Global Geometry
- 3D Reconstruction
- Single-image to 3D
- Multiview 3D
- 3D Recognition
- Classification
- Detection
- Segmentation
- 6D Pose Estimation
- 3D Geometry Processing
- Point Cloud Processing
- Learning-based Mesh Processing
- Part-based 3D Understanding
- Part-based Generative Model
- Zero-shot 3D Understanding
- Mobility
- Human and Hand Pose
- 3D Shape Collection
- Pairwise Correspondence
- Collection Correspondence