Deep Learning for 3D Data

CSE291 (E00) - Fall 2022

	Section 1	Theories of Geometry
9/22, 9/27	Intro, Curve Theory, PDF (2022 ver), Annotated PDF (2021 ver)	overview of the course, logistics, curve theory
9/29	Surface Theory	differential map, normal curvature, principal curvature
10/4	Surface Theory (II)	shape operator, first fundamental form, isometry, fundamental theorem of surfaces
10/6, 10/11	Mesh and Point Cloud	polygonal mesh, point cloud
10/13	Rotation and SO(3)	rotation matrix, euler angle, angle-axis, quaternion
	Section 2	3D Deep Learning
	Section 2.1	3D Reconstruction
10/18, 10/20	Learning-based MVS	learning-based MVS, NeRF
10/25	State-of-the-art Neural 3D Capturing	MVSNeRF, NeRFusion, TensoRF
10/27	Single Image to 3D	EMD, Chamfer, mesh deformation
	Section 2.2	3D Data Understanding
11/1	3D Backbone Networks	Volumetric CNN, PointNet
11/3	3D Detection	Frustum PointNet, PointPillar, VoteNet
11/8	6D Pose Estimation	ICP, Umeyama's method, direct method
11/15	6D Pose Estimation (II)	indirect approach, DenseFusion, PVN3D, NOCS
11/17	Intrinsics-based Analysis	geodesic distance, dijkstra's algorithm for geodesics, learning-based method for geodesics, applications
11/22	Guest lecture by Kaichun Mo	3D affordance, object manipulation
11/24	Thanksgiving	No class
	Section 2.3	Structured 3D Learning
11/29	Deformation Models	surface deformation, space deformation, skeleton skinning
12/1	Mesh Processing	misc mesh processing problems and a brief intro to the course next quarter