NeuRoRA: Neural Robust Rotation Averaging

TL;DR

NeuRoRA introduces a neural network approach for robust rotation averaging that learns noise patterns, detects outliers, and fine-tunes camera orientations more accurately and faster than traditional methods.

Contribution

The paper presents a novel neural network architecture combining outlier detection and orientation refinement for rotation averaging, outperforming conventional robust optimization techniques.

Findings

The neural network achieves higher accuracy than traditional methods.

The approach is significantly faster in processing.

It effectively detects and rectifies noisy measurements.

Abstract

Multiple rotation averaging is an essential task for structure from motion, mapping, and robot navigation. The task is to estimate the absolute orientations of several cameras given some of their noisy relative orientation measurements. The conventional methods for this task seek parameters of the absolute orientations that agree best with the observed noisy measurements according to a robust cost function. These robust cost functions are highly nonlinear and are designed based on certain assumptions about the noise and outlier distributions. In this work, we aim to build a neural network that learns the noise patterns from the data and predict/regress the model parameters from the noisy relative orientations. The proposed network is a combination of two networks: (1) a view-graph cleaning network, which detects outlier edges in the view-graph and rectifies noisy measurements; and (2) a fine-tuning network, which fine-tunes an initialization of absolute orientations bootstrapped from the cleaned graph, in a single step. The proposed combined network is very fast, moreover, being trained on a large number of synthetic graphs, it is more accurate than the conventional iterative optimization methods. Although the idea of replacing robust optimization methods by a graph-based network is demonstrated only for multiple rotation averaging, it could easily be extended to other graph-based geometric problems, for example, pose-graph optimization.