Abstract | ||
---|---|---|
Optical Flow estimation in noisy image sequences re- quires a special denoising strategy. Towards this end we introduce a new tensor-driven anisotropic diffusion scheme which is designed to enhance optical-flow-like spatio- temporal structures. This is achieved by selecting diffusiv- ities in a special manner depending on the eigenvalues of the well known structure tensor. We illustrate how the pro- posed choice differs from edge- and coherence-enhancing anisotropic diffusion. Furthermore we extend a recently dis- covered discretization scheme for anisotropic diffusion to 3D data. An automatic stop criterion to terminate the diffu- sion after a suitable time is given. The performance of the introduced method is examined quantitatively using image sequences with a substantial amount of noise added. |
Year | DOI | Venue |
---|---|---|
2001 | 10.1109/ICCV.2001.937571 | Computer Vision, 2001. ICCV 2001. Proceedings. Eighth IEEE International Conference |
Keywords | Field | DocType |
eigenvalues and eigenfunctions,image sequences,accurate optical flow,anisotropic diffusion,coherence-enhancing anisotropic diffusion,denoising strategy,discretization scheme,eigenvalues,noisy image sequences,optical-flow-like spatio-temporal structures,tensor-driven anisotropic diffusion scheme | Noise reduction,Anisotropic diffusion,Discretization,Pattern recognition,Mathematical analysis,Computer science,Algorithm,Optical flow estimation,Structure tensor,Artificial intelligence,Optical flow,Eigenvalues and eigenvectors | Conference |
Volume | ISBN | Citations |
1 | 0-7695-1143-0 | 20 |
PageRank | References | Authors |
2.31 | 13 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Hagen Spies | 1 | 154 | 15.81 |
Hanno Scharr | 2 | 430 | 37.92 |