Quantitative Assessment of Image Segmentation Quality by Random Walk Relaxation Times
Proceedings of the 31st DAGM Symposium on Pattern Recognition
A Variational Framework for Non-local Image Inpainting
EMMCVPR '09 Proceedings of the 7th International Conference on Energy Minimization Methods in Computer Vision and Pattern Recognition
Diffusion maps for edge-aware image editing
ACM SIGGRAPH Asia 2010 papers
Patch reprojections for Non-Local methods
Signal Processing
A Bias-Variance Approach for the Nonlocal Means
SIAM Journal on Imaging Sciences
| Hi-index | 0.00 |
Nonlocal neighborhood filters are modern and powerful techniques for image and signal denoising. In this paper, we give a probabilistic interpretation and analysis of the method viewed as a random walk on the patch space. We show that the method is intimately connected to the characteristics of diffusion processes, their escape times over potential barriers, and their spectral decomposition. In particular, the eigenstructure of the diffusion operator leads to novel insights on the performance and limitations of the denoising method, as well as a proposal for an improved filtering algorithm.