Название: High-Orders Motion Analysis: Computer Vision Methods
Автор: Yan Sun
Издательство: Springer/Southeast University Press
Год: 2024
Язык: английский
Формат: pdf (true), epub
Размер: 67.1 MB

This book shows how different types of motion can be disambiguated into their components in a richer way than that currently possible in computer vision. Previous research of motion analysis has generally not yet considered the basic nature of higher orders of motion such as acceleration. Hence, this book introduces an approximation of the acceleration field using established optical flow techniques. Further, acceleration is decomposed into radial and tangential based on geometry and propagated as a general motion descriptor; this book shows the capability for differentiating different types of motion both on synthesized data and real image sequences.

Beyond acceleration, the higher orders of motion flow and their continuant parts are investigated for further revealing the chaotic motion fields. Naturally, it is possible to extend this notion further: to detect higher orders of image motion. In this respect, this book shows how jerk and snap can be obtained from image sequences. The derived results on test images and heel strike detection in gait analysis illustrate the ability of higher-order motion, which provide the basis for the following research and applications in the future.

This book has six chapters; Chap. 1 gives the overview and intention of this book. Chapter 2 gives a brief introduction of optical flow, and four benchmark algorithms are compared to demonstrate the advantages and weakness of optical flow algorithms. The acceleration algorithms presented in Chap. 3, include the experimental results on both synthetic and real-world images. In Chap. 4, optical flow is decomposed into higher orders and their constituent parts. Chapter 5 describes and evaluates the methodology of detecting heel strike via higher-order motion. Chapter 6 concludes this book; the analysis and results of higher-order motion descriptor show that they are ripe for further investigation and explore potential future application directions.

We hope that the publication of this book will bring a new perspective to researchers and graduate students in the field of video analysis in computer vision.