Oriented geodesic distance based non-local regularisation approach for optic flow estimation

Optical flow (OF) estimation needs spatial coherence regularisation, due to local image noise and the well-known aperture problem. More recently, OF local-region regularisation has been extended to larger or non-local region of regularisation to further deal with the aperture problem. After careful literature review, it has been determined that the criteria used for deciding the degree of motion coherence can be further improved. For this reason, we propose an oriented geodesic distance based motion regularisation scheme. The proposed approach is particular useful in reducing errors in estimating motions along object boundaries, and recovering motions for nearby objects with similar appearance. Experiment results, compared to leading-edge non-local regularisation schemes, have confirmed the superior performance of the proposed approach.

[1]  Xiaofeng Ren,et al.  Local grouping for optical flow , 2008, 2008 IEEE Conference on Computer Vision and Pattern Recognition.

[2]  Michael J. Black,et al.  The Robust Estimation of Multiple Motions: Parametric and Piecewise-Smooth Flow Fields , 1996, Comput. Vis. Image Underst..

[3]  Edward H. Adelson,et al.  Analysis of Contour Motions , 2006, NIPS.

[4]  Daniel Cremers,et al.  Large displacement optical flow computation withoutwarping , 2009, 2009 IEEE 12th International Conference on Computer Vision.

[5]  Horst Bischof,et al.  Motion estimation with non-local total variation regularization , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[6]  Daniel Freedman,et al.  Illumination-invariant tracking via graph cuts , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[7]  Laurent D. Cohen,et al.  Geodesic Methods in Computer Vision and Graphics , 2010, Found. Trends Comput. Graph. Vis..

[8]  Andrew Blake,et al.  Visual Reconstruction , 1987, Deep Learning for EEG-Based Brain–Computer Interfaces.

[9]  Thomas Brox,et al.  High Accuracy Optical Flow Estimation Based on a Theory for Warping , 2004, ECCV.

[10]  Hans-Hellmut Nagel,et al.  An Investigation of Smoothness Constraints for the Estimation of Displacement Vector Fields from Image Sequences , 1983, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[11]  Takeo Kanade,et al.  An Iterative Image Registration Technique with an Application to Stereo Vision , 1981, IJCAI.

[12]  Michael J. Black,et al.  Secrets of optical flow estimation and their principles , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[13]  Andrew Blake,et al.  Geodesic star convexity for interactive image segmentation , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[14]  Berthold K. P. Horn,et al.  Determining Optical Flow , 1981, Other Conferences.

[15]  Thomas Brox,et al.  Universität Des Saarlandes Fachrichtung 6.1 – Mathematik Highly Accurate Optic Flow Computation with Theoretically Justified Warping Highly Accurate Optic Flow Computation with Theoretically Justified Warping , 2022 .