This course covers recent advances is building digital representations of humans for a variety of applications, such as product design, character animation, and medicine. The focus is on building realistic models of real humans, using measurements.
Topics are organized into 6 modules, building up levels of realism in digital human models.
(1) Introduction. Fast-paced overview of the background in 3D geometry, representations of human shape and motion. Anthropometry. General tips on how to critically read and think about research papers, and write reviews.
(2) Shape and Motion. Kinematic representations. Skeletons and Skinning. 3D scanning. Statistical models. Reduced coordinate models.
(3) Clothing. Representing and assembling garments. Simulating drape. A first look at simulating physics.
(4) Soft Bodies. Finite element models. Dynamics.
(5) Machine Learning. Focus on predicting human shape, motion, and dynamics.
(6) The rest of the story. Topics based on student interest. Examples: Measurement of real humans and clothing; Human anatomy and physiology.
Instructor: Dinesh K. Pai, Computer Science, UBC. Office hours by appointment.
Lectures: MW 10:30-12:00 Dempster 101.
2019W Term 1, classes start September 4, 2019 and end November 27, 2019.
Discussions: We will use Piazza for most communications related to the course.
There are two, equally important, aspects of the course work: (1) Critically reading research papers and (2) Term project.
(1) Reading papers (50%): This is a seminar-style course. After the introductory module each module will consist of a short overview lecture, followed by reading and critiquing recent and seminal papers. The papers are mostly selected from SIGGRAPH conference proceedings but may also include a few from related fields, such as biomechanics, computer vision, and machine learning.
We will use a SIGGRAPH-style review process. Each paper is assigned one primary and two secondary reviewers, who will write short reviews to be shared with the class. Students may indicate, in advance, their preferences for papers to review, but the assignment is made by the instructor. The primary reviewer also gives a 10 minute presentation of the paper in class. Everyone participates in class/piazza discussion.
It is anticipated that each student will write reviews for at least 6 papers and present at least 2 papers in class (the exact number depending on class size). These reviews and presentations will be evaluated both by the instructor and by class peers.
(2) Term project (50%): This will require a hands-on activity; literature reviews are not permitted. Typically the project involves implementing or extending a published paper, or conducting a study. There is considerable flexibility in choosing the term project to match student interests, as long as the project is closely related to digital humans. Students may work individually or in small groups. Students are required to submit a 1-page proposal (10%), a 2-page progress report (10%), a final report (25%), and give a short in class presentation of their results (5%).
In addition to having a passion for learning about digital human models, students are expected to have excelled in one or more undergraduate courses that cover both of these areas:
(1) 3D geometry (e.g., computer graphics, computer vision, or robotics).
(2) Scientific computing (e.g., numerical linear algebra, machine learning, or other courses in applied mathematics).
The term projects will require good software engineering skills to implement a published paper.
If you are unsure of your preparation, please contact the instructor. This is the second offering of this course, and we may make small tweaks as we go along.