Scene Coordinate Regression with Angle-Based Reprojection Loss for Camera Relocalization

Xiaotian Li 1 Xiaotian Li Juha Ylioinas 1 Jakob Verbeek 2 Juho Kannala 3
2 LEAR - Learning and recognition in vision
Inria Grenoble - Rhône-Alpes, LJK - Laboratoire Jean Kuntzmann, INPG - Institut National Polytechnique de Grenoble
Abstract : Image-based camera relocalization is an important problem in computer vision and robotics. Recent works utilize convolutional neural networks (CNNs) to regress for pixels in a query image their corresponding 3D world coordinates in the scene. The final pose is then solved via a RANSAC-based optimization scheme using the predicted coordinates. Usually, the CNN is trained with ground truth scene coordinates, but it has also been shown that the network can discover 3D scene geometry automatically by minimizing single-view reprojection loss. However, due to the deficiencies of reprojection loss, the network needs to be carefully initialized. In this paper, we present a new angle-based reprojection loss which resolves the issues of the original reprojection loss. With this new loss function, the network can be trained without careful initialization, and the system achieves more accurate results. The new loss also enables us to utilize available multi-view constraints, which further improve performance.
Type de document :
Communication dans un congrès
ECCV 2018 - Workshop Geometry Meets Deep Learning, Sep 2018, Munich, Germany. pp.1-16
Liste complète des métadonnées

Littérature citée [20 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-01867143
Contributeur : Thoth Team <>
Soumis le : mardi 2 octobre 2018 - 15:44:00
Dernière modification le : samedi 20 octobre 2018 - 01:08:47

Fichier

li18eccvw.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01867143, version 2

Collections

Citation

Xiaotian Li, Xiaotian Li, Juha Ylioinas, Jakob Verbeek, Juho Kannala. Scene Coordinate Regression with Angle-Based Reprojection Loss for Camera Relocalization. ECCV 2018 - Workshop Geometry Meets Deep Learning, Sep 2018, Munich, Germany. pp.1-16. 〈hal-01867143v2〉

Partager

Métriques

Consultations de la notice

116

Téléchargements de fichiers

45