Qualitative recognition of motion using temporal texture

doi:10.1016/1049-9660(92)90087-J

CVGIP: Image Understanding

Volume 56, Issue 1, July 1992, Pages 78-89

CVGIP: Image Underst...

https://doi.org/10.1016/1049-9660(92)90087-J Get rights and content

Abstract

We describe a method of visual motion recognition applicable to a range of naturally occurring motions that are characterized by spatial and temporal uniformity. The underlying motivation is the observation that, for objects that typically move, it is frequently easier to identify them when they are moving than when they are stationary. Specifically, we show that certain statistical spatial and temporal features that can be derived from approximations to the motion field have invariant properties, and can be used to classify regional activities such as windblown trees, ripples on water, or chaotic fluid flow, that are characterized by complex, nonrigid motion. We refer to the technique as temporal texture analysis in analogy to the techniques developed to classify grayscale textures. This recognition approach contrasts with the reconstructive approach that has typified most prior work on motion. We demonstrate the technique on a number of real-world image sequences containing complex movement. The work has practical application in monitoring and surveillance, and as a component of a sophisticated visual system.

References (57)

R. Bajcsy et al.
Texture gradient as a depth cue
Comput. Graphics Image Process.
(1976)
D.H. Ballard
Generalizing the Hough transform to detect arbitrary shapes
Pattern Recognit.
(1981)
J.E. Cutting
Six tenets for event perception
Cognition
(1981)
I. Dinstein
A new technique for visual motion alarm
Pattern Recogn. Lett.
(1988)
R.L. Kashyap et al.
Texture classification using features derived from random field models
Pattern Recognit. Lett.
(1982)
H.H. Nagel
Displacement vectors derived from second order intensity variations in image sequences
Comput. Vision Pattern Recognit. Image Process.
(1983)
K. Prazdny
Determining the instantaneous direction of motion from optical flow generated by a curvilinear moving observer
Comput. Vision Graphics Image Process.
(1981)
G. Adiv
Inherent ambiguities in recovering 3-D motion and structure from a noisy flow field
P. Anandan et al.
Introducing a smoothness constraint in a matching approach for the computation of optical flow fields
P. Anandan
A computational framework and an algorithm for the measurement of visual motion
Int. J. Comput. Vision
(1989)

C.H. Anderson et al.

Change detection and tracking using pyramid transform techniques

S.T. Barnard et al.

Disparity Analysis of Images

IEEE Trans. PAMI

(1980)

L.E. Baum et al.

A maximization technique occurring in the statistical analysis of probabilistic functions of Markov chains

Ann. Math. Stat.

(1970)

R.C. Bolles

Epipolar Plane Analysis: An Approach to Determining Structure from Motion

P.J. Burt et al.

Object tracking with a moving camera

C.H. Chen

A study of texture classifications using spectral features

H.W. Chun

a representation for temporal sequence and duration in massively parallel networks: Exploiting link connections

R.W. Conners et al.

A theoretical comparison of texture algorithms

IEEE Trans. PAMI

(1980)

J.E. Cutting et al.

Recognizing friends by their walk: Gait perception without familiarity cues

Bull. Psychon. Soc.

(1977)

J.E. Elaman

Finding Structure in Time

J.P. Ewart

Neuroethology of releasing mechanisms: Preycatching in toads

Behav. Brain Sci.

(1987)

J.E. Feldman

Time, Space and Form in Vision

N.H. Goddard

Representing and recognizing event sequences

W.E.L. Grimson

The combinatorics of local constraints in model-based recognition and localization from sparse data

J. ACM

(1986)

R.M. Haralick et al.

Textural features for image classification

IEEE Trans. Systems Man Cybernet.

(Nov. 1973)

D. Heeger

Optical flow from spatio-temporal filters

E.C. Hildreth et al.

The analysis of visual motion from computational theory to neural mechanisms

Ann. Rev. Neurosci.

(1987)

D.D. Hoffman et al.

The interpretation of biological motion

Biol. Cybernet.

(1982)

Cited by (122)

Dynamic texture analysis using networks generated by deterministic partially self-avoiding walks
2020, Physica A: Statistical Mechanics and its Applications
Dynamic textures are sequences of images (video) with the concept of texture patterns extended to the spatiotemporal domain. This research field has attracted attention due to the range of applications in different areas of science and the emergence of a large number of multimedia datasets. Unlike the static textures, the methods for dynamic texture analysis also need to deal with the time domain, which increases the challenge for representation. Thus, it is important to obtain features that properly describe the appearance and motion properties of the dynamic texture. In this paper, we propose a new method for dynamic texture analysis based on Deterministic Partially Self-avoiding Walks (DPSWs) and network science theory. Here, each pixel of the video is considered a vertex of the network and the edges are given by the movements of the deterministic walk between the pixels. The feature vector is obtained by calculating network measures from the networks generated by the DPSWs. The modeled networks incorporate important characteristics of the DPSWs transitivity and their spatial arrangement in the video. In this paper, two different strategies are tested to apply the DPSWs in the video and capture appearance and motion characteristics. One strategy applies the DPSWs in three orthogonal planes and the other is based on spatial and temporal descriptors. We validate our proposed method in Dyntex++ and UCLA (and their variants) databases, which are two well-known dynamic texture databases. The results have demonstrated the effectiveness of the proposed approach using a small feature vector for both strategies. Also, the proposed method improved the performance when compared to the previous DPSW-based method and the network-based method.
Dynamic texture analysis with diffusion in networks
2019, Digital Signal Processing: A Review Journal
Dynamic texture is a field of research that has gained considerable interest from computer vision community due to the explosive growth of multimedia databases. In addition, dynamic texture is present in a wide range of videos, which makes it very important in expert systems based on videos such as medical systems, traffic monitoring systems, forest fire detection system, among others. In this paper, a new method for dynamic texture characterization based on diffusion in directed networks is proposed. The dynamic texture is modeled as a directed network. The method consists in the analysis of the dynamic of this network after a series of graph cut transformations based on the edge weights. For each network transformation, the activity for each vertex is estimated. The activity is the relative frequency that one vertex is visited by random walks in balance. Then, texture descriptor is constructed by concatenating the activity histograms. The main contributions of this paper are the use of directed network modeling and diffusion in network to dynamic texture characterization. These tend to provide better performance in dynamic textures classification. Experiments with rotation and interference of the motion pattern were conducted in order to demonstrate the robustness of the method. The proposed approach is compared to other dynamic texture methods on two very well known dynamic texture database and on traffic condition classification, and outperform in most of the cases.
Data acquisition and simulation of dynamic flame with temperature distribution
2018, Graphical Models
Data acquisition and reuse of dynamic flame provides a new way for realistic flame simulation. The virtual model has a comparable physical data support while reconstructing the realistic flame. In this paper, we focus on the acquisition and simulation of dynamic flame with temperature distribution. Firstly, a synthesis 2D dynamic flame texture method is proposed to simulate and edit the high frequency detail of dynamic flame based on temperature distribution. Secondly, a 3D flame reconstruction method is put forward to reconstruct and edit temperature field from the captured multi-view flame images based on pyrometry and fluid dynamics. The simulated flame has the constraints and support of the physical temperature data; and the dynamic flame can be edited based on the temperature distribution intuitively. Different flame scenes were tested and the experimental results show that our method is effective in flame temperature data acquisition and simulation.
Auroral event representation based on the n-ary fusion of multiple oriented energies
2017, Neurocomputing
The auroral event is defined as a period of various special aurora appearance. It is a good way to understand the interaction among the solar wind plasma, the terrestrial magnetic field, and the Earth’s atmosphere, since the motion of auroral event supplies a powerful tool to discover the underlying coherence among those natural phenomena. Therefore, studies on different types of auroral events have attracted growing attentions. In this study, an auroral event representation approach is proposed for automatic aurora classification. Specifically, 3D steerable filters are used to generate space-time oriented energies, which can simultaneously capture texture and motion of the auroral event. Then, n-ary fusion operator is used to combine oriented energies together to generate a unified representation, which also makes the representation has the property of gray-scale invariance. In order to capture roughly global shape information, a block partition scheme is used to further represent the auroral event. Finally, the performance of the representation is evaluated by the auroral event classification. The superior experimental results demonstrate the effective of our proposed auroral event representation method.
Efficient bat-inspired block matching algorithm with novel motion energy directional histograms for dynamic texture fast recognition
2024, Multimedia Tools and Applications
Fast dynamic texture recognition based on block estimation and axial spatio-temporal motion vector components
2023, Signal, Image and Video Processing

View all citing articles on Scopus

View full text

Regular articleQualitative recognition of motion using temporal texture

Abstract

Comput. Graphics Image Process.

Pattern Recognit.

Cognition

Pattern Recogn. Lett.

Pattern Recognit. Lett.

Comput. Vision Pattern Recognit. Image Process.

Comput. Vision Graphics Image Process.

Inherent ambiguities in recovering 3-D motion and structure from a noisy flow field

Introducing a smoothness constraint in a matching approach for the computation of optical flow fields

A computational framework and an algorithm for the measurement of visual motion

Int. J. Comput. Vision

Change detection and tracking using pyramid transform techniques

Disparity Analysis of Images

IEEE Trans. PAMI

A maximization technique occurring in the statistical analysis of probabilistic functions of Markov chains

Ann. Math. Stat.

Epipolar Plane Analysis: An Approach to Determining Structure from Motion

Object tracking with a moving camera

A study of texture classifications using spectral features

a representation for temporal sequence and duration in massively parallel networks: Exploiting link connections

A theoretical comparison of texture algorithms

IEEE Trans. PAMI

Recognizing friends by their walk: Gait perception without familiarity cues

Bull. Psychon. Soc.

Finding Structure in Time

Neuroethology of releasing mechanisms: Preycatching in toads

Behav. Brain Sci.

Time, Space and Form in Vision

Representing and recognizing event sequences

The combinatorics of local constraints in model-based recognition and localization from sparse data

J. ACM

Textural features for image classification

IEEE Trans. Systems Man Cybernet.

Optical flow from spatio-temporal filters

The analysis of visual motion from computational theory to neural mechanisms

Ann. Rev. Neurosci.

The interpretation of biological motion

Biol. Cybernet.

Regular article
Qualitative recognition of motion using temporal texture