Robustness model fitting method based on supermap mode search

Abstract

A robustness model fitting method based on supermap mode search relates to the computer visual technology; the method comprises the following steps: preparing a dataset; setting up a supermap model G-(V, E), one model assumption corresponds to one peak v of the supermap, and a data point corresponds to a super edge e, so each peak is connected with an inner point of the corresponding model assumption, i.e., the super edge; using a non-parameter kernel density estimation method to evaluate a weight fraction w (v) of each peak v, so a model fitting problem can be converted into a mode search problem in the supermap; providing mode search through searching weight peak in the supermap; determining inner points and parameters of each structure through searched mode and supermap model; cutting an image according to the parameters and inner points of each structure so as to complete model fitting. The robustness model fitting method can alleviate data distribution sensitivity, and the set supermap needs no conversion, and can be directly applied to mode search.

Description

technical field [0001] The invention relates to computer vision technology, in particular to a robust model fitting method based on hypergraph pattern search. Background technique [0002] With the development of science and technology, our life is increasingly inseparable from the development of computer vision theory and algorithms. How to extract effective information from images, that is, computer recognition of images, has a very important impact on the development of computer vision. Parametric models are effective representations of image information, and model fitting methods can effectively estimate appropriate model parameters from observational data. [0003] In the past 10 years, model fitting methods have been widely used in the field of computer vision, such as motion segmentation, image stitching, optical flow calculation, homography estimation, fundamental matrix estimation, etc. Among the model fitting methods, one of the more popular methods is Random Sam...

AI Technical Summary

Problems solved by technology

[0005] The existing model fitting methods still have a lot of problems in dealing with real data: clustering-based fitting methods (such as KF and J-linkage) are sensitive to data distribution, so they are not suitable for dealing with unbalanced data problems

In addition, this type of method cannot effectively deal with data points that are in the handover of the two models

Hypergraph-based fitting methods (such as RCG) often need to map hypergraphs to simple graphs, which will cause information loss

Other methods (such as AKSWH and T-linkage) also have some problems. For example, when AKSWH selects some meaningful model assumptions, it often removes the assumptions in the corresponding real structure, and T-linkage often causes time complexity deviation through hierarchical clustering. High, not suitable for processing large data

Images