1,087 Hits in 4.4 sec

A Dynamic Programming Solution to Bounded Dejittering Problems [article]

Lukas F. Lang
2017 arXiv   pre-print
We propose a dynamic programming solution to image dejittering problems with bounded displacements and obtain efficient algorithms for the removal of line jitter, line pixel jitter, and pixel jitter.  ...  It is straightforward to apply the above framework to (8), (9), (13) , and (14) . Energy Minimisation on Graphs. A more general point of view is to consider (17) on (undirected) graphs.  ...  term sums over all pairs of vertices which are connected via an edge in the graph.  ... 
arXiv:1703.09161v1 fatcat:j6da4l3eazfizlwm7prcrkadyu

Multi-Atlas Segmentation Using Partially Annotated Data: Methods and Annotation Strategies

Lisa Margret Koch, Martin Rajchl, Wenjia Bai, Christian Frederik Baumgartner, Tong Tong, Jonathan Passerat-Palmbach, Paul Aljabar, Daniel Rueckert
2018 IEEE Transactions on Pattern Analysis and Machine Intelligence  
To address this issue, we first re-examine the labelling problem common in many existing approaches and formulate its solution in terms of a Markov Random Field energy minimisation problem on a graph connecting  ...  This work is licensed under a Creative Commons Attribution 3.0 License.  ...  Recently, [35] applied graph-cuts for co-segmentation of pairs of PET and CT images by minimising an MRF energy function which penalises tumour segmentation differences between a PET and CT image of  ... 
doi:10.1109/tpami.2017.2711020 pmid:28841548 fatcat:j5dhmal3knfvvdchi6lfwb53eq


Clément Dechesne, Clément Mallet, Arnaud Le Bris, Valérie Gouet, Alexandre Hervieu
2016 The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences  
The proposed energy is then minimized using a standard graph-cut method (namely QPBO with α-expansion) in order to produce a segmentation map with a controlled level of details.  ...  A probability map is generated through the tree species classification and inserted with the pixel-based features map in an energetical framework.  ...  This classification is then regularised through an energy minimisation. The regularisation, performed with a graph-cut method, produces homogeneous tree species areas with smooth borders.  ... 
doi:10.5194/isprs-archives-xli-b3-207-2016 fatcat:uxqwmuckvrbq3fgs2yfdqugvie

A quadratic programming approach to image labelling

Z. Fu, A. Robles-Kelly
2008 IET Computer Vision  
The cost function to minimise is convex and its discrete version is equivalent up to a constant additive factor to the target function used in discrete Markov Random Field approaches.  ...  In this paper, we present an alternative continuous relaxation approach to image labeling which makes use of a quadratic cost function over the class labels.  ...  To this end, we propose an alternative approach to solving the MRF inference problem which hinges in the minimisation of a continuous cost function making use of a graph regularisation scheme over a manifold  ... 
doi:10.1049/iet-cvi:20080033 fatcat:mdfrk6m46zflvifhgn3ioh26je

Inverse Learning of Symmetries [article]

Mario Wieser, Sonali Parbhoo, Aleksander Wieczorek, Volker Roth
2020 arXiv   pre-print
Our approach is based on the deep information bottleneck in combination with a continuous mutual information regulariser.  ...  Unlike previous methods, we focus on the challenging task of minimising mutual information in continuous domains.  ...  Acknowledgements We would like to thank Anders S. Christensen, Felix A. Faber, Puck van Gerwen, O. Anatole von Lilienfeld, Sebastian M. Keller, Jimmy C.  ... 
arXiv:2002.02782v2 fatcat:o2u4wdqkkrdudgeervkobx6sle

Simple GNN Regularisation for 3D Molecular Property Prediction Beyond [article]

Jonathan Godwin and Michael Schaarschmidt and Alexander Gaunt and Alvaro Sanchez-Gonzalez and Yulia Rubanova and Petar Veličković and James Kirkpatrick and Peter Battaglia
2022 arXiv   pre-print
In this paper we show that simple noise regularisation can be an effective way to address GNN oversmoothing.  ...  Our regulariser applies well-studied methods in simple, straightforward ways which allow even generic architectures to overcome oversmoothing and achieve state of the art results on quantum chemistry tasks  ...  During training, our noise regularisation approach corrupts the input graph's attributes with noise, and adds a per-node noise correction term.  ... 
arXiv:2106.07971v2 fatcat:lqzt4ijrurhcpg4mh46uq6mdci

Weighted Mean Assignment of a Pair of Correspondences Using Optimisation Functions [chapter]

Carlos Francisco Moreno-García, Francesc Serratosa
2014 Lecture Notes in Computer Science  
It is based on an optimisation method that minimises the cost of the correspondence while forcing it (to the most) to be a weighted mean.  ...  Likewise, optimisation in graph matching has been explored to accelerate and improve pattern recognition systems.  ...  Another option is a standard minimisation approach, to reduce the computational time. Standard minimisation approaches aim to find an optimal element that globally minimises a specific function.  ... 
doi:10.1007/978-3-662-44415-3_31 fatcat:ne44ex2ndrgezpeuk34kn5usga


Kazeem Oyeyemi Oyebode, Jules R. Tapamo
2016 Image Analysis and Stereology  
The graph cut energy function includes a parameter that adjusts its data term and smoothness term relative to each other.  ...  To address the problem of trial and error in manual parameter selection, we propose an intuitive and adaptive parameter selection for cell segmentation using graph cut.  ...  ACKNOWLEDGMENT The authors would like to thank Dr Alex Sigal of the KwaZulu-Natal Research Institute for Tuberculosis and HIV (K-RITH), for providing the private dataset used in this research.  ... 
doi:10.5566/ias.1333 fatcat:55lkcarcyna6bkc56i5df4wywm

Computational shape optimisation for a gradient-enhanced continuum damage model

Fabian Guhr, Leon Sprave, Franz-Joseph Barthold, Andreas Menzel
2020 Computational Mechanics  
An isotropic gradient-enhanced damage model is applied to shape optimisation in order to establish a computational optimal design framework in view of optimal damage distributions.  ...  This approach eliminates the necessity of introducing a local variable, leaving only the global set of equations to be iteratively solved.  ...  At each time step, the sum of an elastic energy and a Griffith-type dissipated energy is minimised.  ... 
doi:10.1007/s00466-019-01810-3 fatcat:4ygspy67xfbpxljmk5rnqqs25m

From Semi-supervised to Transfer Counting of Crowds

Chen Change Loy, Shaogang Gong, Tao Xiang
2013 2013 IEEE International Conference on Computer Vision  
Extensive experiments validate the effectiveness of our approach.  ...  In this study, we propose to address this problem from three perspectives: (1) Instead of exhaustively annotating every single frame, the most informative frames are selected for annotation automatically  ...  This formulation builds on the Laplacian regularised least squares concept [25] , but is reformulated carefully to employ Hessian energy [18, 24] for manifold regularisation due to the latter's superior  ... 
doi:10.1109/iccv.2013.270 dblp:conf/iccv/LoyGX13 fatcat:65fnu4irxjdlnn5lf4wqge2ic4

Feature Selection: An Assessment of Some Evolving Methodologies

A. Abdul Rasheed
2021 Turkish Journal of Computer and Mathematics Education  
This article analyses such methods like LASSO, Boruta, Recursive Feature Elimination (RFE), Regularised Random Forest (RRF) and DALEX.  ...  The dataset of variant sizes is considered to assess the importance of feature selection out of the available features.  ...  An automatic breadth searching and attention searching adjustment approaches to further speedup randomized wrapper-based feature selection was proposed by this research.  ... 
doi:10.17762/turcomat.v12i2.1802 fatcat:tjey2qqfw5ashoqaopnsnl6mia

Proximal approach to denoising hyperspectral images under mixed-noise model

Hazique Aetesam, Kumari Poonam, Suman Kumar Maji
2020 IET Image Processing  
The authors present a proximal approach to hyperspectral image denoising adapted to the mixed noise behaviour of hyperspectral data; named hyperspectral image proximal denoiser (HSIProxDenoiser).  ...  Hence, including both regularisation terms can help achieve the desired denoising performance.  ...  Energy functionals of the class defined by (9) can be minimised using primal dual minimisation algorithms such as PDHG [33] .  ... 
doi:10.1049/iet-ipr.2019.1763 fatcat:fm3wttchqrgj3fulxkmho7xobe

Fitting multidimensional data using gradient penalties and the sparse grid combination technique

Jochen Garcke, Markus Hegland
2009 Computing  
Sparse grids, combined with gradient penalties provide an attractive tool for regularised least squares fitting.  ...  We argue that this is due to the irregular and random data distribution, as well as the proportion of the number of data to the grid resolution.  ...  It uses a gradient based regularisation and discretises the minimisation problem with an independent grid with associated local basis functions.  ... 
doi:10.1007/s00607-009-0027-x fatcat:3zn7mmz4r5b6llzehhnot54ntu

Scan-Less Line Field Optical Coherence Tomography, with Automatic Image Segmentation, as a Measurement Tool for Automotive Coatings

Samuel Lawman, Bryan Williams, Jinke Zhang, Yao-Chun Shen, Yalin Zheng
2017 Applied Sciences  
Recently, graph search algorithms have emerged as a popular [41] [42] [43] [44] approach  ...  To reliably extract layer thicknesses from such images of automotive paint systems, we present an automatic graph search image segmentation algorithm.  ...  The founding sponsors had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, and in the decision to publish the results.  ... 
doi:10.3390/app7040351 fatcat:2u7m6vq5jjhhjfl6iko4gsisrm

Graph-Based Regularization of Binary Classifiers for Texture Segmentation [chapter]

Cyrille Faucheux, Julien Olivier, Romuald Boné
2013 Lecture Notes in Computer Science  
Segmentation is expressed by an indicator function whose quality is evaluated by an energy according to two criteria: the regularity and the fidelity to the data.  ...  It is based on a recent non-local image segmentation approach [Ta, 2009] whose particularity is to represent images with similarity graphs.  ...  de segmentation par régularisation de graphe.  ... 
doi:10.1007/978-3-642-40261-6_37 fatcat:x4esaupdhbetzg2eer6mcpshsm
« Previous Showing results 1 — 15 out of 1,087 results