On Efficient and Robust Metrics for RANSAC Hypotheses and 3D Rigid Registration

TL;DR

This paper introduces new efficient and robust metrics for evaluating RANSAC hypotheses, significantly improving 3D rigid registration accuracy and speed across various scenarios by analyzing inliers and outliers contributions.

Contribution

It proposes novel metrics for RANSAC hypotheses that are faster and more robust, addressing limitations of existing methods in 3D registration tasks.

Findings

Proposed metrics outperform state-of-the-art in accuracy and robustness.

Metrics are computationally efficient and adaptable to different scenarios.

Analysis reveals not all inliers are equal, influencing metric design.

Abstract

This paper focuses on developing efficient and robust evaluation metrics for RANSAC hypotheses to achieve accurate 3D rigid registration. Estimating six-degree-of-freedom (6-DoF) pose from feature correspondences remains a popular approach to 3D rigid registration, where random sample consensus (RANSAC) is a de-facto choice to this problem. However, existing metrics for RANSAC hypotheses are either time-consuming or sensitive to common nuisances, parameter variations, and different application scenarios, resulting in performance deterioration in overall registration accuracy and speed. We alleviate this problem by first analyzing the contributions of inliers and outliers, and then proposing several efficient and robust metrics with different designing motivations for RANSAC hypotheses. Comparative experiments on four standard datasets with different nuisances and application scenarios verify that the proposed metrics can significantly improve the registration performance and are more robust than several state-of-the-art competitors, making them good gifts to practical applications. This work also draws an interesting conclusion, i.e., not all inliers are equal while all outliers should be equal, which may shed new light on this research problem.