3-D Motion Estimation, Understanding, and Prediction from Noisy Image Sequences

Research output: Contribution to journalArticle

Abstract

This paper presents an approach to understanding general 3-D motion of a rigid body from image sequences. Based on dynamics, a locally constant angular momentum (LCAM) model is introduced. The model is local in the sense that it is applied to a limited number of image frames at a time. Specifically, the model constrains the motion, over a local frame subsequence, to be a superposition of precession and translation. Thus, the instantaneous rotation axis of the object is allowed to change through the subsequence. The trajectory of the rotation center is approximated by a vector polynomial. The parameters of the model evolve in time so that they can adapt to long term changes in motion characteristics. The nature and parameters of short term motion can be estimated continuously with the goal of understanding motion through the image sequence. The estimation algorithm presented in this paper is linear, i.e., the algorithm consists of solving simultaneous linear equations. Based on the assumption that the motion is smooth, object positions and motion in the near future can be predicted, and short missing subsequences can be recovered. Noise smoothing is achieved by overdetermination and a leastsquares criterion. The framework is flexible in the sense that it allows both overdetermination in number of feature points and the number of image frames. The number of frames from which the model is derived can be varied according to the complexity of motion and the noise level so as to obtain stable and good estimates of parameters over the entire image sequence. Simulation results are given for noisy synthetic data and images taken of a model airplane.

Original languageEnglish (US)
Pages (from-to)370-389
Number of pages20
JournalIEEE transactions on pattern analysis and machine intelligence
VolumePAMI-9
Issue number3
DOIs
StatePublished - May 1987

Fingerprint

Motion Estimation
Motion estimation
Image Sequence
3D
Motion
Prediction
Subsequence
Angular momentum
Model
Linear equations
Simultaneous equations
Feature Point
Noisy Data
Term
Synthetic Data
Estimation Algorithms
Angular Momentum
Trajectories
Aircraft
Rigid Body

Keywords

  • Computer vision
  • analysis
  • dynamic model
  • image sequence
  • motion
  • motion estimation
  • motion prediction
  • motion under-
  • standing

ASJC Scopus subject areas

  • Software
  • Computer Vision and Pattern Recognition
  • Computational Theory and Mathematics
  • Artificial Intelligence
  • Applied Mathematics

Cite this

@article{df70a1887ff549019edd406def291a92,
title = "3-D Motion Estimation, Understanding, and Prediction from Noisy Image Sequences",
abstract = "This paper presents an approach to understanding general 3-D motion of a rigid body from image sequences. Based on dynamics, a locally constant angular momentum (LCAM) model is introduced. The model is local in the sense that it is applied to a limited number of image frames at a time. Specifically, the model constrains the motion, over a local frame subsequence, to be a superposition of precession and translation. Thus, the instantaneous rotation axis of the object is allowed to change through the subsequence. The trajectory of the rotation center is approximated by a vector polynomial. The parameters of the model evolve in time so that they can adapt to long term changes in motion characteristics. The nature and parameters of short term motion can be estimated continuously with the goal of understanding motion through the image sequence. The estimation algorithm presented in this paper is linear, i.e., the algorithm consists of solving simultaneous linear equations. Based on the assumption that the motion is smooth, object positions and motion in the near future can be predicted, and short missing subsequences can be recovered. Noise smoothing is achieved by overdetermination and a leastsquares criterion. The framework is flexible in the sense that it allows both overdetermination in number of feature points and the number of image frames. The number of frames from which the model is derived can be varied according to the complexity of motion and the noise level so as to obtain stable and good estimates of parameters over the entire image sequence. Simulation results are given for noisy synthetic data and images taken of a model airplane.",
keywords = "Computer vision, analysis, dynamic model, image sequence, motion, motion estimation, motion prediction, motion under-, standing",
author = "Juyang Weng and Huang, {Thomas S} and Narendra Ahuja",
year = "1987",
month = "5",
doi = "10.1109/TPAMI.1987.4767920",
language = "English (US)",
volume = "PAMI-9",
pages = "370--389",
journal = "IEEE Transactions on Pattern Analysis and Machine Intelligence",
issn = "0162-8828",
publisher = "IEEE Computer Society",
number = "3",

}

TY - JOUR

T1 - 3-D Motion Estimation, Understanding, and Prediction from Noisy Image Sequences

AU - Weng, Juyang

AU - Huang, Thomas S

AU - Ahuja, Narendra

PY - 1987/5

Y1 - 1987/5

N2 - This paper presents an approach to understanding general 3-D motion of a rigid body from image sequences. Based on dynamics, a locally constant angular momentum (LCAM) model is introduced. The model is local in the sense that it is applied to a limited number of image frames at a time. Specifically, the model constrains the motion, over a local frame subsequence, to be a superposition of precession and translation. Thus, the instantaneous rotation axis of the object is allowed to change through the subsequence. The trajectory of the rotation center is approximated by a vector polynomial. The parameters of the model evolve in time so that they can adapt to long term changes in motion characteristics. The nature and parameters of short term motion can be estimated continuously with the goal of understanding motion through the image sequence. The estimation algorithm presented in this paper is linear, i.e., the algorithm consists of solving simultaneous linear equations. Based on the assumption that the motion is smooth, object positions and motion in the near future can be predicted, and short missing subsequences can be recovered. Noise smoothing is achieved by overdetermination and a leastsquares criterion. The framework is flexible in the sense that it allows both overdetermination in number of feature points and the number of image frames. The number of frames from which the model is derived can be varied according to the complexity of motion and the noise level so as to obtain stable and good estimates of parameters over the entire image sequence. Simulation results are given for noisy synthetic data and images taken of a model airplane.

AB - This paper presents an approach to understanding general 3-D motion of a rigid body from image sequences. Based on dynamics, a locally constant angular momentum (LCAM) model is introduced. The model is local in the sense that it is applied to a limited number of image frames at a time. Specifically, the model constrains the motion, over a local frame subsequence, to be a superposition of precession and translation. Thus, the instantaneous rotation axis of the object is allowed to change through the subsequence. The trajectory of the rotation center is approximated by a vector polynomial. The parameters of the model evolve in time so that they can adapt to long term changes in motion characteristics. The nature and parameters of short term motion can be estimated continuously with the goal of understanding motion through the image sequence. The estimation algorithm presented in this paper is linear, i.e., the algorithm consists of solving simultaneous linear equations. Based on the assumption that the motion is smooth, object positions and motion in the near future can be predicted, and short missing subsequences can be recovered. Noise smoothing is achieved by overdetermination and a leastsquares criterion. The framework is flexible in the sense that it allows both overdetermination in number of feature points and the number of image frames. The number of frames from which the model is derived can be varied according to the complexity of motion and the noise level so as to obtain stable and good estimates of parameters over the entire image sequence. Simulation results are given for noisy synthetic data and images taken of a model airplane.

KW - Computer vision

KW - analysis

KW - dynamic model

KW - image sequence

KW - motion

KW - motion estimation

KW - motion prediction

KW - motion under-

KW - standing

UR - http://www.scopus.com/inward/record.url?scp=0023349052&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0023349052&partnerID=8YFLogxK

U2 - 10.1109/TPAMI.1987.4767920

DO - 10.1109/TPAMI.1987.4767920

M3 - Article

AN - SCOPUS:0023349052

VL - PAMI-9

SP - 370

EP - 389

JO - IEEE Transactions on Pattern Analysis and Machine Intelligence

JF - IEEE Transactions on Pattern Analysis and Machine Intelligence

SN - 0162-8828

IS - 3

ER -