Robust Diffusion Kernels for Optical Flowsmoothing
Authors:
- 1 September 2006
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE) in 2007 IEEE Workshop on Machine Learning for Signal Processing
- p. 415-420
Abstract
This paper provides a comparison study among a set of novel algorithms that implement robust diffusion on optical flows. The proposed algorithms combine the anisotropic smoothing ability of the heat kernel and the outlier rejection mechanism of robust statistics algorithms. The diffusion kernel is considered Gaussian, where the covariance matrix is the local Hessian. This enables the kernel to detect significant transitions in the signal. In this study we show that diffusion does not eliminate outliers but rather spreads them around. We calculate the resulting bias induced by diffusing the outliers in their neighbourhood. On the other hand robust statistics operators reject the outliers from the diffusion process. Alpha-trimmed mean and median statistics are considered in combination with the diffusion processing. The proposed algorithms are applied for smoothing optical flow.Keywords
This publication has 8 references indexed in Scilit:
- Vector-valued image regularization with PDEs: a common framework for different applicationsIEEE Transactions on Pattern Analysis and Machine Intelligence, 2005
- Stochastic differential equations and geometric flowsIEEE Transactions on Image Processing, 2002
- Smoothing low-SNR molecular images via anisotropic median-diffusionIEEE Transactions on Medical Imaging, 2002
- Prediction and tracking of moving objects in image sequencesIEEE Transactions on Image Processing, 2000
- Optical flow estimation and moving object segmentation based on median radial basis function networkIEEE Transactions on Image Processing, 1998
- Robust anisotropic diffusionIEEE Transactions on Image Processing, 1998
- Performance of optical flow techniquesInternational Journal of Computer Vision, 1994
- Scale-space and edge detection using anisotropic diffusionIEEE Transactions on Pattern Analysis and Machine Intelligence, 1990