Image sequence segmentation using 3-D structure tensor and curve evolution

Zhang, J.; Gao, Jianbo; Liu, W.

doi:10.1109/76.920192

Cited by 31 publications

(26 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The gradient structure tensor method (GSTM) is a newer approach 1,[4][5][6][7] . With the rapid advance of computer technology, the GSTM has been employed for real-time motion estimation in recent years 15,16 .…”

Section: The Gradient Structure Tensor Methodsmentioning

confidence: 99%

“…Motion estimation techniques in the spatial domain may be classified as being either gradient-based or correlationbased methods (also commonly referred to as block matching or template matching) 1 . Gradient-based techniques can be further divided into spatio-temporal gradient methods (often simply referred to as gradient methods) [1][2][3] and gradient structure tensor methods (also referred to as gradient square tensor methods or 3-d structure tensor methods) 1,[4][5][6][7] . Gradient-based methods are in general used to obtain a dense optical flow field or motion vectors.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Untitled

Kondo¹,

Kongprawechnon²

2009

ScienceAsia

View full text Add to dashboard Cite

Changing lighting conditions cause temporal variations of image intensities and make most existing motion estimation techniques ineffective. As a solution to this problem, we use gradient orientation information, which depends very little on changes of image intensities, in place of commonly used image features such as intensities and gradients. By employing gradient orientation, two conventional motion estimation techniques (the spatio-temporal gradient method and the gradient structure tensor method) can be transformed into methods that are far more robust to changing image intensities. Simulation results on both synthetic and real image sequences show that the proposed methods perform motion estimation remarkably well irrespective of time-varying image intensities. In addition, the proposed methods also cope better with the aperture problem in which only unidirectional gradients are available and large erroneous motion estimates are produced.

show abstract

Section: The Gradient Structure Tensor Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Untitled

Kondo¹,

Kongprawechnon²

2009

ScienceAsia

View full text Add to dashboard Cite

show abstract

“…spatial only versus temporal). (5) To resolve this ambiguity and to classify the video regions experiencing motion, the eigenvalues and the associated eigenvectors of J are usually analyzed [10], [11]. However eigenvalue decompositions at every pixel is computationally expensive especially if real time performance is required.…”

Section: A 3d Structure Tensorsmentioning

confidence: 99%

“…10, the elements of the flux tensor incorporate information about temporal gradient changes which leads to efficient discrimination between stationary and moving image features. Thus the trace of the flux tensor matrix which can be compactly written and computed as, (11) and can be directly used to classify moving and non-moving regions without the need for expensive eigenvalue decompositions. If motion vectors are needed then Eq.…”

Section: B Flux Tensorsmentioning

confidence: 99%

Flux Tensor Constrained Geodesic Active Contours with Sensor Fusion for Persistent Object Tracking

Bunyak

Palaniappan²,

Nath³

et al. 2007

JMM

View full text Add to dashboard Cite

This paper makes new contributions in motion detection, object segmentation and trajectory estimation to create a successful object tracking system. A new efficient motion detection algorithm referred to as the flux tensor is used to detect moving objects in infrared video without requiring background modeling or contour extraction. The flux tensor-based motion detector when applied to infrared video is more accurate than thresholding "hot-spots", and is insensitive to shadows as well as illumination changes in the visible channel. In real world monitoring tasks fusing scene information from multiple sensors and sources is a useful core mechanism to deal with complex scenes, lighting conditions and environmental variables. The object segmentation algorithm uses level set-based geodesic active contour evolution that incorporates the fusion of visible color and infrared edge informations in a novel manner. Touching or overlapping objects are further refined during the segmentation process using an appropriate shape-based model. Multiple object tracking using correspondence graphs is extended to handle groups of objects and occlusion events by Kalman filter-based cluster trajectory analysis and watershed segmentation. The proposed object tracking algorithm was successfully tested on several difficult outdoor multispectral videos from stationary sensors and is not confounded by shadows or illumination variations.

show abstract

“…Recently, Zhang and his coworkers (see Zhang et al 114,115 and Gao 113 ) proposed an object segmentation in motion imagery that consisted of: (1) fast global motion compensation, (2) robust frame differencing, and (3) level set based curve evolution. The algorithm segments the objects in the temporal sequence with a moving background using the fast global motion estimation.…”

Section: Motion Segmentation Using Frame Difference Via Pde (Zhang/uw )mentioning

confidence: 99%

Modeling Segmentation via Geometric Deformable Regularizers, Pde and Level Sets in Still and Motion Imagery: A Revisit

Suri

Reden

et al. 2001

Int. J. Image Grap.

View full text Add to dashboard Cite

Partial Differential Equations (PDEs) have dominated image processing research recently. The three main reasons for their success are: first, their ability to transform a segmentation modeling problem into a partial differential equation framework and their ability to embed and integrate different regularizers into these models; second, their ability to solve PDEs in the level set framework using finite difference methods; and third, their easy extension to a higher dimensional space. This paper is an attempt to survey and understand the power of PDEs to incorporate into geometric deformable models for segmentation of objects in 2D and 3D in still and motion imagery. The paper first presents PDEs and their solutions applied to image diffusion. The main concentration of this paper is to demonstrate the usage of regularizers in PDEs and level set framework to achieve the image segmentation in still and motion imagery. Lastly, we cover miscellaneous applications such as: mathematical morphology, computation of missing boundaries for shape recovery and low pass filtering, all under the PDE framework. The paper concludes with the merits and the demerits of PDEs and level set-based framework for segmentation modeling. The paper presents a variety of examples covering both synthetic and real world images.

show abstract

Image sequence segmentation using 3-D structure tensor and curve evolution

Cited by 31 publications

References 45 publications

Untitled

Untitled

Flux Tensor Constrained Geodesic Active Contours with Sensor Fusion for Persistent Object Tracking

Modeling Segmentation via Geometric Deformable Regularizers, Pde and Level Sets in Still and Motion Imagery: A Revisit

Contact Info

Product

Resources

About