FSI Schemes: Fast Semi-Iterative Solvers for PDEs and Optimisation Methods
Many tasks in image processing and computer vision are modelled by diffusion processes, variational formulations, or constrained optimisation problems. Basic iterative solvers such as explicit schemes, Richardson iterations, or projected gradient descent methods are simple to implement and well-suited for parallel computing. However, their efficiency suffers from severe step size restrictions. As a remedy we introduce a simple and highly efficient acceleration strategy, leading to so-called Fast Semi-Iterative (FSI) schemes that extrapolate the basic solver iteration with the previous iterate. To derive suitable extrapolation parameters, we establish a recursion relation that connects box filtering with an explicit scheme for 1D homogeneous diffusion. FSI schemes avoid the main drawbacks of recent Fast Explicit Diffusion (FED) and Fast Jacobi techniques, and they have an interesting connection to the heavy ball method in optimisation. Our experiments show their benefits for anisotropic diffusion inpainting, nonsmooth regularisation, and Nesterov’s worst case problems for convex and strongly convex optimisation.
KeywordsElliptic Problem Time Step Size Explicit Scheme Linear Diffusion Diffusion Evolution
Our research has been partially funded by the Deutsche Forschungsgemeinschaft (DFG) through a Gottfried Wilhelm Leibniz Prize for Joachim Weickert and by the Graduate School of Computer Science at Saarland University. This is gratefully acknowledged.
- 8.Iijima, T.: Basic theory on normalization of pattern (in case of typical one-dimensional pattern). Bull. Electrotech. Lab. 26, 368–388 (1962). in JapaneseGoogle Scholar
- 15.Rumelhart, D.E., Hinton, G.E., Williams, R.J.: Learning internal representations by error propagation, chap. 8. In: Rumelhart, D.E., McClelland, J.L. (eds.) Parallel Distributed Processing: Explorations in the Microstructure of Cognition, vol. 1, pp. 318–362. MIT Press, Cambridge (1986)Google Scholar
- 18.Sutskever, I., Martens, J., Dahl, G., Hinton, G.: On the importance of initialization and momentum in deep learning. In: Proceedings of 30th International Conference on Machine Learning, pp. 1139–1147. Atlanta, GA, June 2013Google Scholar
- 25.Chzhao-Din, Y.: Some difference schemes for the solution of the first boundary value problem for linear differential equations with partial derivatives. Ph.D. thesis, Moscow State University (1958). in RussianGoogle Scholar