Scene Coordinate Regression with Angle-Based Reprojection Loss for Camera Relocalization - Archive ouverte HAL Access content directly
Conference Papers Year :

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

(1) , (1) , (2) , (3)
1
2
3
Xiaotian Li
  • Function : Author
Juha Ylioinas
  • Function : Author
Juho Kannala
  • Function : Author
  • PersonId : 947888

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.
Fichier principal
Vignette du fichier
li18eccvw.pdf (677.93 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-01867143 , version 1 (04-09-2018)
hal-01867143 , version 2 (02-10-2018)

Identifiers

Cite

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.229-245, ⟨10.1007/978-3-030-11015-4_19⟩. ⟨hal-01867143v2⟩
369 View
541 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More