Learning to Solve Hard Minimal Problems

TL;DR

This paper introduces a learning-based method to efficiently solve complex minimal geometric problems in RANSAC, significantly reducing spurious solutions and computation time.

Contribution

It proposes a novel learning strategy for selecting initial problem-solution pairs, enabling faster and more accurate solutions to minimal geometric problems.

Findings

Solved relative pose of three cameras in under 70 microseconds.

Reduced computation time compared to traditional methods.

Benchmarking on two-view pose problem demonstrates effectiveness.

Abstract

We present an approach to solving hard geometric optimization problems in the RANSAC framework. The hard minimal problems arise from relaxing the original geometric optimization problem into a minimal problem with many spurious solutions. Our approach avoids computing large numbers of spurious solutions. We design a learning strategy for selecting a starting problem-solution pair that can be numerically continued to the problem and the solution of interest. We demonstrate our approach by developing a RANSAC solver for the problem of computing the relative pose of three calibrated cameras, via a minimal relaxation using four points in each view. On average, we can solve a single problem in under 70 $\mu s.$ We also benchmark and study our engineering choices on the very familiar problem of computing the relative pose of two calibrated cameras, via the minimal case of five points in two views.