This book discusses mathematical foundations of statistical inference for building a 3-D model of the environment from image and sensor data that contain noise - a central task for autonomous robots guided by video cameras and sensors. A theoretical accuracy bound is derived for the optimization procedure for maximizing the reliability of the estimation based on noisy data, and practical computational schemes that attain that bound are derived. Many synthetic and real data examples are given to demonstrate that conventional methods are not optimal and how accuracy improves if truly optimal methods are employed.
Author(s): Kenichi Kanatani (Eds.)
Series: Machine intelligence and pattern recognition 18
Publisher: Elsevier
Year: 1996
Language: English
Pages: 1-509
City: Amsterdam; New York
Content:
Preface
Pages v-vi
Kenichi Kanatani
Chapter 1 Introduction Original Research Article
Pages 1-26
Chapter 2 Fundamentals of linear algebra Original Research Article
Pages 27-60
Chapter 3 Probabilities and statistical estimation Original Research Article
Pages 61-93
Chapter 4 Representation of geometric objects Original Research Article
Pages 95-130
Chapter 5 Geometric correction Original Research Article
Pages 131-170
Chapter 6 3-D computation by stereo vision Original Research Article
Pages 171-207
Chapter 7 Parametric fitting Original Research Article
Pages 209-246
Chapter 8 Optimal filter Original Research Article
Pages 247-265
Chapter 9 Renormalization Original Research Article
Pages 267-294
Chapter 10 Applications of geometric estimation Original Research Article
Pages 295-323
Chapter 11 3-D motion analysis Original Research Article
Pages 325-368
Chapter 12 3-D interpretation of optical flow Original Research Article
Pages 369-414
Chapter 13 Information criterion for model selection Original Research Article
Pages 415-450
Chapter 14 General theory of geometric estimation Original Research Article
Pages 451-499
Index
Pages 501-509
