Making minimal solvers fast

Abstract

In this paper we propose methods for speeding up

A Practical Rank-Constrained Eight-Point Algorithm for Fundamental Matrix Estimation

“…In fact, we are only interested in the real eigenvalues (and their corresponding eigenvectors) in a narrow interval, rather than a full factorization. It is expected that our algorithm could be significantly accelerated by using the characteristic polynomial based technique in [4]. We leave the exploration of this acceleration technique as our future work.…”

“…In our experiments, we use three image pairs from the Oxford multiview geometry datasets 4 . Specifically, the pairs in Fig.1(a), Fig.1(b) and Fig.1(c) are from the Arial Views I, Dinosaur and Merton College III sequence, representing different motion types such as (near) translation, turntable motion and backward-forward motion.…”

Efficient Solvers for Minimal Problems by Syzygy-Based Reduction

“…While our focus has been on constructing smaller templates there has been a number of works that have addressed the problem of making solvers more numerically stable [9,26] and faster [6,32]. It is possible that these methods could be applied in conjunction with our method, but this is left as further work.…”

Distortion Varieties

“…Let us now analyze two-view geometry for the model (5). The quantity λ = µ/f 2 is our distortion parameter.…”