Learning Scene Geometry for Visual Localization in Challenging Conditions
Résumé
We propose a new approach for outdoor large scale image based localization that can deal with challenging scenarios like cross-season, cross-weather, day/night and long-term localization. The key component of our method is a new learned global image descriptor, that can effectively benefit from scene geometry information during training. At test time, our system is capable of inferring the depth map related to the query image and use it to increase localization accuracy. We are able to increase recall@1 performances by 2.15% on cross-weather and long-term localization scenario and by 4.24% points on a challenging winter/summer localization sequence versus state-of-the-art methods. Our method can also use weakly annotated data to localize night images across a reference dataset of daytime images.
Mots clés
image colour analysis
learning (artificial intelligence)
image retrieval
Robots
Image reconstruction
Geometry
feature extraction
Visualization
Training
Decoding
object detection
robot vision
daytime images
learning scene geometry
visual localization
outdoor large scale image
cross-season
learned global image descriptor
scene geometry information
depth map
query image
localization accuracy
cross-weather
long-term localization scenario
night images
winter localization sequence
summer localization sequence
| Origine | Fichiers produits par l'(les) auteur(s) |
|---|
Loading...