You are previewing Computer Vision for Visual Effects.
O'Reilly logo
Computer Vision for Visual Effects

Book Description

Modern blockbuster movies seamlessly introduce impossible characters and action into real-world settings using digital visual effects. These effects are made possible by research from the field of computer vision, the study of how to automatically understand images. Computer Vision for Visual Effects will educate students, engineers and researchers about the fundamental computer vision principles and state-of-the-art algorithms used to create cutting-edge visual effects for movies and television. The author describes classical computer vision algorithms used on a regular basis in Hollywood (such as blue screen matting, structure from motion, optical flow and feature tracking) and exciting recent developments that form the basis for future effects (such as natural image matting, multi-image compositing, image retargeting and view synthesis). He also discusses the technologies behind motion capture and three-dimensional data acquisition. More than 200 original images demonstrating principles, algorithms and results, along with in-depth interviews with Hollywood visual effects artists, tie the mathematical concepts to real-world filmmaking.

Table of Contents

  1. Cover Page
  2. Title Page
  3. Copyright Page
  4. 1 Introduction
    1. 1.1 Computer Vision for Visual Effects
    2. 1.2 This Book’s Organization
    3. 1.3 Background and Prerequisites
    4. 1.4 Acknowledgments
  5. 2 Image Matting
    1. 2.1 Matting Terminology
    2. 2.2 Blue-Screen, Green-Screen, and Difference Matting
    3. 2.3 Bayesian Matting
    4. 2.4 Closed-Form Matting
    5. 2.5 Markov Random Fields for Matting
    6. 2.6 Random-Walk Methods
    7. 2.7 Poisson Matting
    8. 2.8 Hard-Segmentation-Based Matting
    9. 2.9 Video Matting
    10. 2.10 Matting Extensions
    11. 2.11 Industry Perspectives
    12. 2.12 Notes and Extensions
    13. 2.13 Homework Problems
  6. 3 Image Compositing and Editing
    1. 3.1 Compositing Hard-Edged Pieces
    2. 3.2 Poisson Image Editing
    3. 3.3 Graph-Cut Compositing
    4. 3.4 Image Inpainting
    5. 3.5 Image Retargeting and Recompositing
    6. 3.6 Video Recompositing, Inpainting, and Retargeting
    7. 3.7 Industry Perspectives
    8. 3.8 Notes and Extensions
    9. 3.9 Homework Problems
  7. 4 Features and Matching
    1. 4.1 Feature Detectors
    2. 4.2 Feature Descriptors
    3. 4.3 Evaluating Detectors and Descriptors
    4. 4.4 Color Detectors and Descriptors
    5. 4.5 Artificial Markers
    6. 4.6 Industry Perspectives
    7. 4.7 Notes and Extensions
    8. 4.8 Homework Problems
  8. 5 Dense Correspondence and Its Applications
    1. 5.1 Affine and Projective Transformations
    2. 5.2 Scattered Data Interpolation
    3. 5.3 Optical Flow
    4. 5.4 Epipolar Geometry
    5. 5.5 Stereo Correspondence
    6. 5.6 Video Matching
    7. 5.7 Morphing
    8. 5.8 View Synthesis
    9. 5.9 Industry Perspectives
    10. 5.10 Notes and Extensions
    11. 5.11 Homework Problems
  9. 6 Matchmoving
    1. 6.1 Feature Tracking for Matchmoving
    2. 6.2 Camera Parameters and Image Formation
    3. 6.3 Single-Camera Calibration
    4. 6.4 Stereo Rig Calibration
    5. 6.5 Image Sequence Calibration
    6. 6.6 Extensions of Matchmoving
    7. 6.7 Industry Perspectives
    8. 6.8 Notes and Extensions
    9. 6.9 Homework Problems
  10. 7 Motion Capture
    1. 7.1 The Motion Capture Environment
    2. 7.2 Marker Acquisition and Cleanup
    3. 7.3 Forward Kinematics and Pose Parameterization
    4. 7.4 Inverse Kinematics
    5. 7.5 Motion Editing
    6. 7.6 Facial Motion Capture
    7. 7.7 Markerless Motion Capture
    8. 7.8 Industry Perspectives
    9. 7.9 Notes and Extensions
    10. 7.10 Homework Problems
  11. 8 Three-Dimensional Data Acquisition
    1. 8.1 Light Detection and Ranging (LiDAR)
    2. 8.2 Structured Light Scanning
    3. 8.3 Multi-View Stereo
    4. 8.4 RegisteringD Datasets
    5. 8.5 Industry Perspectives
    6. 8.6 Notes and Extensions
    7. 8.7 Homework Problems
  12. A Optimization Algorithms for Computer Vision
    1. A.1 Dynamic Programming
    2. A.2 Belief Propagation
    3. A.3 Graph Cuts and á-Expansion
    4. A.4 Newton Methods for Nonlinear Sum-of-Squares Optimization
  13. B Figure Acknowledgments
  14. Bibliography
  15. Index