Fast surface interpolation using hierarchical basis functions

By Richard Szeliski. 1989

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Optimization & Training

Citations in the corpus

Deep Residual Learning for Image Recognition (2015, background): “It has been shown [3, 45, 46] that these solvers converge much faster than standard solvers that are unaware of the residual nature of the solutions.”
The Fast Bilateral Solver (2015, methodology): “We will present a novel preconditioner which is similar in spirit to hierarchical basis functions [33] or push-pull interpolation [12], but adapted to our task using the bilateral pyramid techniques presented in [2].”
On 3D scene flow and structure estimation (2001, methodology): “As suggested in [11] and [16], an iterative relaxation method or conjugate gradient method may be used to minimize Eq.”
Modelling with implicit surfaces that interpolate (2002, methodology): “This approach has been widely used for height-field interpolation and deformable models, and examples of its use can be found in [29, 27, 7, 31].”
Locally adapted hierarchical basis preconditioning (2006, methodology): “First, the choice of basis functions and the number of levels is problem-dependent [Szeliski 1990].”
Fourier Analysis of the 2D Screened Poisson Equation for Gradient Domain Problems (2008, methodology): “A comparison of memory and run-time performance of our method (FS) to other state-of-the-art methods (QT [5], out-of-core SM [6], HB [21], LHAB) [27].”
Beyond pixels: exploring new representations and applications for motion analysis (2009, methodology): “The user can also select these modes, and even choose to produce a smooth flow field interpolated using the preconditioned conjugate gradient algorithm described in [112].”
ImageNet classification with deep convolutional neural networks (2012, unknown): “An alternative to Multigrid is hierarchical basis pre-conditioning [45, 46], which relies on variables that represent residual vectors between two scales.”
Finding nemo: Deformable object class modelling using curve matching (2010, background): “Each image is of a different class instance, therefore identification of point correspondences is often impossible.”