 |
15. 3D Models From Photos and Videos
Monday, Tutorial, 10:30 am - 12:15 pm
Room 14 A-B
Level: Intermediate
How 3D models can be obtained from photos or video acquired with a hand-held camera. The approach is based on advanced automatic techniques that avoid camera calibration and a priori scene knowledge, and that gradually retrieve more and more information about the images, the cameras, and the scene.
Prerequisites
Basic understanding of 3D geometry and perspective projection. A full appreciation of the course requires some knowledge of projective geometry, computer vision, and image-processing techniques.
Topics
Feature extraction, feature tracking, (wide-baseline) feature matching, multi-view relations, projective structure and motion recovery, self-calibration, bundle adjustment, image-pair rectification, dense stereo matching, multi-view matching, 3D surface modeling and texturing, image-based rendering, and applications.
Organizer
Marc Pollefeys
University of North Carolina at Chapel Hill
Lecturers
Luc Van Gool
Katholieke Universiteit Leuven/Eidgenössische Technische Hochschule Zürich
Marc Pollefeys
University of North Carolina at Chapel Hill
Schedule
| 10:30 |
Introduction
Pollefeys |
| 10:45 |
Feature Tracking and Matching
Van Gool |
| 11:10 |
Structure and Motion Recovery
Pollefeys |
| 11:30 |
Dense Model Reconstruction
Pollefeys |
| Noon |
Examples, Applications, and Questions
Pollefeys and Van Gool |
|
|
|
|
|
