HSCNet++ : Hierarchical Scene Coordinate Classification and Regression for Visual Localization with Transformer

Loading...
Thumbnail Image

Access rights

openAccess
publishedVersion

URL

Journal Title

Journal ISSN

Volume Title

A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä

Major/Subject

Mcode

Degree programme

Language

en

Pages

21

Series

International Journal of Computer Vision, Volume 132, issue 7, pp. 2530-2550

Abstract

Visual localization is critical to many applications in computer vision and robotics. To address single-image RGB localization, state-of-the-art feature-based methods match local descriptors between a query image and a pre-built 3D model. Recently, deep neural networks have been exploited to regress the mapping between raw pixels and 3D coordinates in the scene, and thus the matching is implicitly performed by the forward pass through the network. However, in a large and ambiguous environment, learning such a regression task directly can be difficult for a single network. In this work, we present a new hierarchical scene coordinate network to predict pixel scene coordinates in a coarse-to-fine manner from a single RGB image. The proposed method, which is an extension of HSCNet, allows us to train compact models which scale robustly to large environments. It sets a new state-of-the-art for single-image localization on the 7-Scenes, 12-Scenes, Cambridge Landmarks datasets, and the combined indoor scenes.

Description

Other note

Citation

Wang, S, Laskar, Z, Melekhov, I, Li, X, Zhao, Y, Tolias, G & Kannala, J 2024, 'HSCNet++ : Hierarchical Scene Coordinate Classification and Regression for Visual Localization with Transformer', International Journal of Computer Vision, vol. 132, no. 7, pp. 2530-2550. https://doi.org/10.1007/s11263-023-01982-9