6D Camera Relocalization in Ambiguous Scenes via Continuous Multimodal Inference

TL;DR

This paper introduces a multimodal camera relocalization method that models ambiguities using continuous mixture models on the pose manifold, predicting multiple hypotheses and uncertainties to improve localization in ambiguous scenes.

Contribution

It proposes a novel end-to-end deep neural network framework that uses Bingham and Gaussian distributions for pose estimation, effectively capturing scene ambiguities and uncertainties.

Findings

Outperforms existing methods on ambiguous scenes

Successfully models multiple pose hypotheses and uncertainties

Demonstrates robustness on synthetic and real datasets

Abstract

We present a multimodal camera relocalization framework that captures ambiguities and uncertainties with continuous mixture models defined on the manifold of camera poses. In highly ambiguous environments, which can easily arise due to symmetries and repetitive structures in the scene, computing one plausible solution (what most state-of-the-art methods currently regress) may not be sufficient. Instead we predict multiple camera pose hypotheses as well as the respective uncertainty for each prediction. Towards this aim, we use Bingham distributions, to model the orientation of the camera pose, and a multivariate Gaussian to model the position, with an end-to-end deep neural network. By incorporating a Winner-Takes-All training scheme, we finally obtain a mixture model that is well suited for explaining ambiguities in the scene, yet does not suffer from mode collapse, a common problem with mixture density networks. We introduce a new dataset specifically designed to foster camera localization research in ambiguous environments and exhaustively evaluate our method on synthetic as well as real data on both ambiguous scenes and on non-ambiguous benchmark datasets. We plan to release our code and dataset under $\href{https://multimodal3dvision.github.io}{multimodal3dvision.github.io}$.