Improved 3D Point-Line Mapping Regression for Camera Relocalization

TL;DR

This paper introduces a novel architecture for camera re-localization that independently learns 3D point and line features, improving mapping accuracy and efficiency over previous combined or feature-matching methods.

Contribution

It proposes a new independent feature learning architecture for 3D point-line mapping, enhancing accuracy and reducing overfitting in camera relocalization.

Findings

Significant improvement in 3D map point and line regression accuracy.

Independent feature learning reduces overfitting and improves localization.

Method outperforms previous approaches in experimental evaluations.

Abstract

In this paper, we present a new approach for improving 3D point and line mapping regression for camera re-localization. Previous methods typically rely on feature matching (FM) with stored descriptors or use a single network to encode both points and lines. While FM-based methods perform well in large-scale environments, they become computationally expensive with a growing number of mapping points and lines. Conversely, approaches that learn to encode mapping features within a single network reduce memory footprint but are prone to overfitting, as they may capture unnecessary correlations between points and lines. We propose that these features should be learned independently, each with a distinct focus, to achieve optimal accuracy. To this end, we introduce a new architecture that learns to prioritize each feature independently before combining them for localization. Experimental results demonstrate that our approach significantly enhances the 3D map point and line regression performance for camera re-localization. The implementation of our method will be publicly available at: https://github.com/ais-lab/pl2map/.