RaP-Net: A Region-wise and Point-wise Weighting Network to Extract Robust Features for Indoor Localization
This work provides a solution for more robust indoor visual localization, which is crucial for applications like robotics and augmented reality, by improving feature extraction in challenging indoor environments.
This paper addresses the challenge of unreliable feature extraction in indoor visual localization caused by dynamic objects and repetitive regions. The authors propose RaP-Net, a network that predicts both region-wise invariability and point-wise reliability to extract more robust features, leading to significantly improved performance in indoor localization compared to state-of-the-art methods.
Feature extraction plays an important role in visual localization. Unreliable features on dynamic objects or repetitive regions will interfere with feature matching and challenge indoor localization greatly. To address the problem, we propose a novel network, RaP-Net, to simultaneously predict region-wise invariability and point-wise reliability, and then extract features by considering both of them. We also introduce a new dataset, named OpenLORIS-Location, to train the proposed network. The dataset contains 1553 images from 93 indoor locations. Various appearance changes between images of the same location are included and can help the model to learn the invariability in typical indoor scenes. Experimental results show that the proposed RaP-Net trained with OpenLORIS-Location dataset achieves excellent performance in the feature matching task and significantly outperforms state-of-the-arts feature algorithms in indoor localization. The RaP-Net code and dataset are available at https://github.com/ivipsourcecode/RaP-Net.