Objects Matter: Learning Object Relation Graph for Robust Camera Relocalization

TL;DR

This paper introduces a novel object relation graph to enhance feature distinctiveness in camera relocalization, significantly improving robustness in complex environments by leveraging semantic and spatial object relationships.

Contribution

The paper proposes a deep object relation graph module integrated into pose regression models to improve relocalization accuracy under challenging conditions.

Findings

Significant performance improvements on indoor and outdoor datasets.

Outperforms previous relocalization methods.

Enhances robustness in environments with appearance changes.

Abstract

Visual relocalization aims to estimate the pose of a camera from one or more images. In recent years deep learning based pose regression methods have attracted many attentions. They feature predicting the absolute poses without relying on any prior built maps or stored images, making the relocalization very efficient. However, robust relocalization under environments with complex appearance changes and real dynamics remains very challenging. In this paper, we propose to enhance the distinctiveness of the image features by extracting the deep relationship among objects. In particular, we extract objects in the image and construct a deep object relation graph (ORG) to incorporate the semantic connections and relative spatial clues of the objects. We integrate our ORG module into several popular pose regression models. Extensive experiments on various public indoor and outdoor datasets demonstrate that our method improves the performance significantly and outperforms the previous approaches.