S3E-GNN: Sparse Spatial Scene Embedding with Graph Neural Networks for Camera Relocalization

TL;DR

S3E-GNN introduces a novel end-to-end learning framework utilizing graph neural networks for efficient and accurate camera relocalization, outperforming traditional methods in challenging environments.

Contribution

The paper presents a new learning-based approach combining scene embedding and GNNs for robust camera relocalization, enhancing accuracy over traditional techniques.

Findings

Outperforms Bag-of-words in relocalization accuracy

Effective in challenging environments

Utilizes scene embedding and graph neural networks

Abstract

Camera relocalization is the key component of simultaneous localization and mapping (SLAM) systems. This paper proposes a learning-based approach, named Sparse Spatial Scene Embedding with Graph Neural Networks (S3E-GNN), as an end-to-end framework for efficient and robust camera relocalization. S3E-GNN consists of two modules. In the encoding module, a trained S3E network encodes RGB images into embedding codes to implicitly represent spatial and semantic embedding code. With embedding codes and the associated poses obtained from a SLAM system, each image is represented as a graph node in a pose graph. In the GNN query module, the pose graph is transformed to form a embedding-aggregated reference graph for camera relocalization. We collect various scene datasets in the challenging environments to perform experiments. Our results demonstrate that S3E-GNN method outperforms the traditional Bag-of-words (BoW) for camera relocalization due to learning-based embedding and GNN powered scene matching mechanism.