 |
21. Projectors: Advanced Geometric Issues in Applications
Monday, Tutorial, 10:30 am - 12:15 pm
Room 33 A-C
Level: Intermediate
The latest computer vision and graphics techniques for projector-based displays. In addition to design, rendering, and calibration algorithms in conventional projector-based systems, this course provides a practical guide to relevant topics in image warping, projective geometry, transfer functions for curved screens, camera-assisted methods, and projector-based augmented reality.
Prerequisites
General knowledge of basic computer graphics theory and practice, including perspective projection and rendering. Also helpful: familiarity with video projection and basic concepts in computer vision. The primary intended audiences are designers and programmers who use or build applications for innovative projector-based systems.
Topics
Calibration and projective geometry: pin-hole projector model, homography, curved transfer, global consistency, estimating pose, radial distortion, depth extraction. Image warping: rendering techniques, exploiting graphics hardware, intensity blending functions. Parametric and non-parametric approaches. Integration of camera, tracker, mirrors. Applications: casually aligned displays, spatially augmented reality, shadow removal.
Organizer and Lecturer
Ramesh Raskar
Mitsubishi Electric Research Laboratories (MERL)
Schedule
| 10:30 |
Introduction
- Projection Model
Raskar |
| 10:40 |
Calibration
- Parameter Estimation and Robustness
- Image Transfer Functions, Planar and Quadric
- Non-parametric Approaches
Raskar |
| 11:10 |
Geometric Relationship
- Device Integration, Trackers, and Mirrors
Raskar |
| 11:30 |
Rendering
- Image Generation and Warping
- Multi-Projector Seamless Displays
- Projector-Based Augmented Reality
Raskar |
| 12:05 |
Examples and Discussion |
|
|
|
|
|
