Velocity tuned generalized Sobel operators for multiresolution computation of optical flow

Gokstorp, M.; Danielsson, Per-Erik

doi:10.1109/icip.1994.413674

Cited by 5 publications

(3 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There is an upper bound for valid velocity estimates at each level, thus larger velocities must be estimated at levels of lower resolution. The connection between spatial frequency and maximum speed to be detected is given in [7].…”

Section: System Overviewmentioning

confidence: 99%

Dense motion field estimation using spatial filtering and quasi eigenfunction approximations

Kristoffersen

Austvoll

Engan

2005

IEEE International Conference on Image Processing 2005

View full text Add to dashboard Cite

Earlier research on a particular phase based method for estimation of dense motion fields show promising results for complicated sequences. Typical for phase based methods is a decomposition of the sequence into directional signals by directional filters. From each directional signal a component velocity can be estimated. In this context we propose a method for estimating component velocities which further improves the accuracy, by modelling 2D spacetime slices in the direction of the filters as single component AM-FM modulated signals. A method for demodulation based on quasi eigenfunction approximation is the basis for estimation of component velocities, along with a tensor formulation and eigenvector analysis of this tensor. We demonstrate the method on the well known reference sequence Yosemite fly-by and the Rubic's cube sequence.

show abstract

Section: System Overviewmentioning

confidence: 99%

Dense motion field estimation using spatial filtering and quasi eigenfunction approximations

Kristoffersen

Austvoll

Engan

2005

IEEE International Conference on Image Processing 2005

View full text Add to dashboard Cite

show abstract

“…In the current implementation the kernel size is a mapping of the marker size bounded by the feasible kernel range that is between 3×3 and 9×9. We perform non-linear gradient approximation using generalized separable Sobel kernels [11, 12] for fast edge detection and smoothing. In polynomial transform representation we have…”

Section: Interactive Segmentationmentioning

confidence: 99%

Interactive segmentation for geographic atrophy in retinal fundus images

Lee

Rjh

Laine

2008

2008 42nd Asilomar Conference on Signals, Systems and Computers

View full text Add to dashboard Cite

Fundus auto-fluorescence (FAF) imaging is a non-invasive technique for in vivo ophthalmoscopic inspection of age-related macular degeneration (AMD), the most common cause of blindness in developed countries. Geographic atrophy (GA) is an advanced form of AMD and accounts for 12–21% of severe visual loss in this disorder [3]. Automatic quantification of GA is important for determining disease progression and facilitating clinical diagnosis of AMD. The problem of automatic segmentation of pathological images still remains an unsolved problem. In this paper we leverage the watershed transform and generalized non-linear gradient operators for interactive segmentation and present an intuitive and simple approach for geographic atrophy segmentation. We compare our approach with the state of the art random walker [5] algorithm for interactive segmentation using ROC statistics. Quantitative evaluation experiments on 100 FAF images show a mean sensitivity/specificity of 98.3/97.7% for our approach and a mean sensitivity/specificity of 88.2/96.6% for the random walker algorithm.

show abstract

“…Horn and Schunck solve the aperture problem by assuming that the flow should be as smooth as possible. Methods based on spatiotemporal filtering with velocity tuned filters [7,10,[14][15][16] assume that the flow is constant over the support of their filters. Block matching methods [3,9] rely on the assumption that the motion is constant over small windows of the picture.…”

Section: (T; X(t)) = I (0; X(0))mentioning

confidence: 99%

Discrete Wavelet Analysis for Fast Optic Flow Computation

Bernard¹

2001

Applied and Computational Harmonic Analysis

View full text Add to dashboard Cite

This paper describes an optic flow estimation method based on a discrete wavelet basis analysis. The differential optic flow equation is projected onto analytic wavelets. This gives small local linear systems (3-5 equations) that are solved to find the visual displacement. In this way, we solve the problems of time aliasing and aperture. Since the coefficients of the systems can be computed with filter banks, the estimation of a flow map costs O(N) operations (if one image of the sequence has N pixels). Our method also measures illumination changes. A convergence theorem is also stated and proved. 

show abstract

Velocity tuned generalized Sobel operators for multiresolution computation of optical flow

Cited by 5 publications

References 5 publications

Dense motion field estimation using spatial filtering and quasi eigenfunction approximations

Dense motion field estimation using spatial filtering and quasi eigenfunction approximations

Interactive segmentation for geographic atrophy in retinal fundus images

Discrete Wavelet Analysis for Fast Optic Flow Computation

Contact Info

Product

Resources

About