Räumlich-zeitliche Berechnung des optischen Flusses mit nichtlinearen flußabhängigen Glattheitstermen

Informatik aktuellPublished 1 January 1999

Joachim Weickert, Christoph Schnörr

DOI10.1007/978-3-642-60243-6_37

Source URL

Citations9

SJR score0.19

SNIP0.16

TL;DR

Jede Form impliziter Glattung, sei es durch einen Vorverarbeitungsschritt zur Rauschunterdruckung oder bei der Schatzung partieller Ableitungen der Bildfunktion, wurde vermieden.

Abstract

Nicht quadratische Variationsansätze sind die mathematische Grundlage leistungsfähiger Algorithmen zur Berechnung stückweise glatter optischer Flußfelder in digitalen Bildfolgen. In dem vorliegenden Artikel erweitern wir diese Ansätze zu nichtlinearen und flußabhängigen Funktionalen, die über einem räumlich-zeitlichen Intervall definiert sind. Um den Effekt dieser Erweiterung studieren zu können, basiert unsere Implementierung ausschließlich auf dem entsprechenden System von Diffusions-Reaktions-Gleichungen. Jede Form impliziter Glättung, sei es durch einen Vorverarbeitungsschritt zur Rauschunterdrückung oder bei der Schätzung partieller Ableitungen der Bildfunktion, wurde vermieden. Numerische Ergebnisse für Realwelt bildfolgen zeigen, daß unser räumlich-zeitlicher Ansatz (i) den berechneten optischen Fluß signifikant verbessert, (ii) Hintergrundrauschen stark unterdrückt und (iii) Bewegungsgrenzen deutlicher hervortreten läßt. Der zusätzliche Berechnungsaufwand macht weniger als einen Faktor 2 gegenüber einem rein räumlichen Ansatz aus.

Keywords

Biochemistry, Genetics and Molecular BiologyEngineering

IEEE Transactions on Pattern Analysis and Machine IntelligenceScale-space and edge detection using anisotropic diffusion

12,040 Citations1990Pietro Perona, Jitendra Malik

Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE<title>Determining Optical Flow</title>

7,496 Citations1981Berthold K. P. Horn, Brian G. Schunck

An iterative implementation is shown which successfully computes the optical flow for a number of synthetic image sequences and is robust in that it can handle image sequences that are quantized rather coarsely in space and time.

International Journal of Computer VisionPerformance of optical flow techniques

4,271 Citations1994John A. Barron, David J. Fleet +1 more

These comparisons are primarily empirical, and concentrate on the accuracy, reliability, and density of the velocity measurements; they show that performance can differ significantly among the techniques the authors implemented.

IEEE Transactions on Medical ImagingNonlinear anisotropic filtering of MRI data

1,144 Citations1992Guido Gerig, O. Kübler +2 more

In contrast to acquisition-based noise reduction methods a postprocess based on anisotropic diffusion is proposed, which overcomes the major drawbacks of conventional filter methods, namely the blurring of object boundaries and the suppression of fine structural details.

International Journal of Computer VisionComputation of component image velocity from local phase information

1,074 Citations1990David J. Fleet, Allan D. Jepson

The resulting technique is predominantly linear, efficient, and suitable for parallel processing, and is local in space-time, robust with respect to noise, and permits multiple estimates within a single neighborhood.

IEEE Transactions on Pattern Analysis and Machine IntelligenceAn Investigation of Smoothness Constraints for the Estimation of Displacement Vector Fields from Image Sequences

766 Citations1986Hans-Hellmut Nagel, W. Enkelmann

Results are reported of an investigation about how an ``oriented smoothness'' constraint may be formulated and evaluated about how variations of the displacement vector field only in directions with small or no variation of gray values are restricted.

IEEE Transactions on Pattern Analysis and Machine IntelligenceScene Segmentation from Visual Motion Using Global Optimization

300 Citations1987David W. Murray, Bernard F. Buxton

To compute the flow predicted by the segmentation, a recent method for reconstructing the motion and orientation of planar surface facets is used and the search for the globally optimal segmentation is performed using simulated annealing.

International Journal of Computer VisionComputation and analysis of image motion: A synopsis of current problems and methods

250 Citations1996Amar Mitiche, Patrick Bouthémy

A structured synopsis of the problems in image motion computation and analysis, and of the methods proposed, exposing the underlying models and supporting assumptions are offered.

Robust dynamic motion estimation over time

241 Citations2002Michael J. Black, P. Anandan

A highly parallel incremental stochastic minimization algorithm is presented which has a number of advantages over previous approaches and the incremental nature of the scheme makes it dynamic and permits the detection of occlusion and disocclusion boundaries.

International Journal of Computer VisionRobust computation of optical flow in a multi-scale differential framework

226 Citations1995Joseph Weber, Jitendra Malik

A new algorithm for computing optical flow in a differential framework based on a robust version of total least squares is developed, incorporating only past time frames.