Computer Vision: Camera Calibration & 3D Reconstruction (Do, 15.01.2015)

Anmeldung erforderlich

RWTH

Für RWTH-Angehörige und aus dem RWTH-Netz verfügbar

Anmelden
  • Einbetten

Kapitel:

00:00:00
Lecture 18: Camera Calibration & 3D Reconstruction
00:00:55
Course Outline
00:01:32
Recap: What Is Stereo Vision?
00:02:03
Recap: Depth with Stereo - Basic Idea
00:03:27
Recap: Epipolar Geometry
00:05:32
Recap: Stereo Geometry With Calibrated Cameras
00:08:28
Recap: Essential Matrix
00:11:22
Recap: Essential Matrix and Epipolar Lines
00:12:16
Recap: Stereo Image Rectification
00:15:29
Recap: Stereo Reconstruction
00:16:12
Correspondence Problem
00:16:57
Dense Correspondence Search
00:19:13
Example: Window Search
00:23:13
Effect of Window Size
00:24:52
Alternative: Sparse Correspondence Search
00:27:28
Dense vs. Sparse
00:30:43
Difficulties in Similarity Constraint
00:31:27
Possible Sources of Error?
00:32:17
Summary: Stereo Reconstruction
00:34:23
Recap: A General Point (SVD)
00:35:49
Topics of This Lecture
00:36:08
Recall: Pinhole Camera Model
00:39:53
Pinhole Camera Model
00:41:15
Camera Coordinate System
00:42:22
Principal Point Offset
00:45:13
Pixel Coordinates: Non-Square Pixels
00:46:50
Camera Rotation and Translation
00:51:10
Summary: Camera Parameters
00:57:00
Camera Parameters: Degrees of Freedom
01:03:11
Calibrating a Camera
01:04:15
Camera Calibration
01:05:08
Camera Calibration: Obtaining the Points
01:08:59
Camera Calibration: DLT Algorithm
01:16:30
Camera Calibration
01:17:01
Camera Calibration: Some Practical Tips
01:20:16
Topics of This Lecture
01:20:27
Two-View Geometry
01:22:03
Revisiting Triangulation
01:23:53
Triangulation: 1) Geometric Approach
01:25:04
Triangulation: 2) Linear Algebraic Approach
01:28:25
Triangulation: 3) Nonlinear Approach
01:31:29
References and Further Reading