Energy-Based Geometric Multi-model Fitting

  title={Energy-Based Geometric Multi-model Fitting},
  author={Hossam N. Isack and Yuri Boykov},
  journal={International Journal of Computer Vision},
Geometric model fitting is a typical chicken-&-egg problem: data points should be clustered based on geometric proximity to models whose unknown parameters must be estimated at the same time. Most existing methods, including generalizations of RANSAC, greedily search for models with most inliers (within a threshold) ignoring overall classification of points. We formulate geometric multi-model fitting as an optimal labeling problem with a global energy function balancing geometric errors and… 

Interacting Geometric Priors For Robust Multimodel Fitting

A novel energy with high-level geometric priors that consider interactions between geometric models, such that certain preferred model configurations may be induced, which is useful in dealing with unknown distributions of data errors and outliers.

Geometric Multi-model Fitting with a Convex Relaxation Algorithm

A novel method for fitting multiple geometric models to multi-structural data via convex relaxation that results in an energy minimisation that is as much as two orders of magnitude faster on comparable architectures thus bringing real-time, robust performance to a wider set of geometric multi-model fitting problems.

Energy Based Multi-Model Fitting and Matching Problems

This thesis introduces global objective functions for multi-model fitting using various forms of regularization (spatial smoothness and sparsity) and proposes a graph-cut based optimization algorithm, PEaRL.

Energy based multiple model fitting for non-rigid structure from motion

This paper reformulate the 3D reconstruction of deformable surfaces from monocular video sequences as a labeling problem in which a set of labels, instead of a single one, is assigned to each variable and proposes a mathematical formulation of this new model and shows how it can be efficiently optimized with a variant of α-expansion.

Joint optimization of fitting & matching in multi-view reconstruction

This paper proposes several fit-&-match energy formulations based on a generalization of the assignment problem and developed an efficient solver based on min-cost-max-flow algorithm that finds near optimal solutions.

Energy Based Multi-model Fitting & Matching for 3D Reconstruction

This work jointly solves feature matching and multi-model fitting problems by optimizing one energy, based on the generalization of the assignment problem and its efficient min-cost-max-flow solver.

Multi-Class Model Fitting by Energy Minimization and Mode-Seeking

Considering that a group of outliers may form spatially coherent structures in the data, a cross-validation-based technique removing statistically insignificant instances is proposed, which achieves fast and robust optimization.

G2MF-WA: Geometric multi-model fitting with weakly annotated data

This paper proposes a novel method to make full use of weak annotating data to boost multi-model fitting performance and produces noticeably better results than state-of-the-art techniques in most cases.

Multi-model fitting based on minimum spanning tree

A novel approach to the computation of primitive geometrical structures, where no prior knowledge about the visual scene is available and a high level of noise is expected, based on the grouping principles of proximity and similarity, of points and preliminary models is presented.

Interacting Geometric Priors For Robust Multi-Model Fitting (Supplementary Material)

It is impossible to penalise the geometric inconsistency using (6) because the smoothness cost is set to zero (i.e., l = 0), and in the applications where the smooths assumption is not valid (e.g., vanishing point detection), it is impossible.



Fast Approximate Energy Minimization with Label Costs

The main algorithmic contribution is an extension of α-expansion that also optimizes “label costs” with well-characterized optimality bounds, which has a natural interpretation as minimizing description length (MDL) and sheds light on classical algorithms like K-means and expectation-maximization (EM).

Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography

New results are derived on the minimum number of landmarks needed to obtain a solution, and algorithms are presented for computing these minimum-landmark solutions in closed form that provide the basis for an automatic system that can solve the Location Determination Problem under difficult viewing.

Accelerated Hypothesis Generation for Multi-structure Robust Fitting

This work proposes a fundamentally new approach to accelerate hypothesis sampling by guiding it with information derived from residual sorting that is naturally capable of handling data with multiple model instances and excels in applications which easily frustrate other techniques.

A polynomial-time bound for matching and registration with outliers

This work presents a framework for computing optimal transformations, aligning one point set to another, in the presence of outliers, and develops several algorithms which make efficient use of convex programming.

Fast approximate energy minimization via graph cuts

This paper proposes two algorithms that use graph cuts to compute a local minimum even when very large moves are allowed, and generates a labeling such that there is no expansion move that decreases the energy.

Two-view multibody structure-and-motion with outliers through model selection

Multibody structure-and-motion (MSaM) is the problem in establishing the multiple-view geometry of several views of a 3D scene taken at different times, where the scene consists of multiple rigid

Two-View Motion Segmentation from Linear Programming Relaxation

  • Hongdong Li
  • Computer Science
    2007 IEEE Conference on Computer Vision and Pattern Recognition
  • 2007
A new mixture-of-fundamental-matrices model is proposed to describe the multibody motions from two views that can be naturally formulated as a linear programming (LP) problem and can be solved efficiently by linear program relaxation.

Optimal consensus set for digital line and plane fitting

By using a digital model instead of a continuous one, it is shown that it can generate all possible consensus sets for model fitting and efficiently searches the optimal solution with time complexity O(Nd log N) for dimension d.

MLESAC: A New Robust Estimator with Application to Estimating Image Geometry

A new robust estimator MLESAC is presented which is a generalization of the RANSAC estimator which adopts the same sampling strategy as RANSac to generate putative solutions, but chooses the solution that maximizes the likelihood rather than just the number of inliers.

Region Competition: Unifying Snakes, Region Growing, and Bayes/MDL for Multiband Image Segmentation

A novel statistical and variational approach to image segmentation based on a new algorithm, named region competition, derived by minimizing a generalized Bayes/minimum description length (MDL) criterion using the variational principle is presented.