GeoPose3K: Mountain landscape dataset for camera pose estimation in outdoor environments

作者：

Highlights：

• New dataset of images with known camera poses in mountainous terrain is introduced.

• The images in the dataset are accompanied by additional synthetic data and images.

• A semi-automatic approach for the dataset acquisition is proposed.

• Baseline evaluation of visual localization and orientation estimation is presented.

摘要

•New dataset of images with known camera poses in mountainous terrain is introduced.•The images in the dataset are accompanied by additional synthetic data and images.•A semi-automatic approach for the dataset acquisition is proposed.•Baseline evaluation of visual localization and orientation estimation is presented.

论文关键词：Camera pose estimation,Visual geo-localization,Camera orientation estimation,Image-to-model registration,Digital terrain models,Semantic segmentation

论文评审过程：Received 18 August 2016, Revised 10 February 2017, Accepted 13 May 2017, Available online 19 June 2017, Version of Record 22 September 2017.

论文官网地址：https://doi.org/10.1016/j.imavis.2017.05.009