# MultiLink: Multi-class Structure Recovery via Agglomerative Clustering and Model Selection

Source: [MultiLink: Multi-class Structure Recovery via Agglomerative Clustering and Model Selection](<https://www.sota2.com/research/paper/68fa789e1e209131cea79605>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-12-22T11:45:19.906Z

Authors: Luca Magri, Filippo Leveni, Giacomo Boracchi

Published: 2025-05-16T05:32:02.000Z

## Abstract

We address the problem of recovering multiple structures of different classes in a dataset contaminated by noise and outliers. In particular, we consider geometric structures defined by a mixture of underlying parametric models (e.g. planes and cylinders, homographies and fundamental matrices), and we tackle the robust fitting problem by preference analysis and clustering. We present a new algorithm, termed MultiLink, that simultaneously deals with multiple classes of models. MultiLink combines on-the-fly model fitting and model selection in a novel linkage scheme that determines whether two clusters are to be merged. The resulting method features many practical advantages with respect to methods based on preference analysis, being faster, less sensitive to the inlier threshold, and able to compensate limitations deriving from hypotheses sampling. Experiments on several public datasets demonstrate that Multi-Link favourably compares with state of the art alternatives, both in multi-class and single-class problems. Code is publicly made available for download.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Homography fitting on AdelaideRMF Homographies 19 scenes](<https://www.sota2.com/research/sota/homography-fitting-on-adelaidermf-homographies-19-scenes.md>): Homography fitting; AdelaideRMF Homographies 19 scenes; Avg Misclassification Error: 5.5.

- [Two-view motion fitting on AdelaideRMF Two-view motions 19 scenes](<https://www.sota2.com/research/sota/two-view-motion-fitting-on-adelaidermf-two-view-motions-19-scenes.md>): Two-view motion fitting; AdelaideRMF Two-view motions 19 scenes; Avg. Misclassification Error: 8.6.

- [Motion Segmentation on Hopkins Motions 155 scenes](<https://www.sota2.com/research/sota/motion-segmentation-on-hopkins-motions-155-scenes.md>): Motion Segmentation; Hopkins Motions 155 scenes; Runtime (s): not recorded.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2505.10874>)

[PDF](<https://arxiv.org/pdf/2505.10874.pdf>)

[Implementation](<https://github.com/magrilu/multilink.git>)