Benchmarking 6DOF Outdoor Visual Localization in Changing Conditions

@article{Sattler2018Benchmarking6O,
  title={Benchmarking 6DOF Outdoor Visual Localization in Changing Conditions},
  author={Torsten Sattler and William P. Maddern and Carl Toft and Akihiko Torii and Lars Hammarstrand and Erik Stenborg and Daniel Safari and M. Okutomi and Marc Pollefeys and Josef Sivic and Fredrik Kahl and Tom{\'a}s Pajdla},
  journal={2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  year={2018},
  pages={8601-8610}
}
Visual localization enables autonomous vehicles to navigate in their surroundings and augmented reality applications to link virtual to real worlds. Practical visual localization approaches need to be robust to a wide variety of viewing condition, including day-night changes, as well as weather and seasonal variations, while providing highly accurate 6 degree-of-freedom (6DOF) camera pose estimates. In this paper, we introduce the first benchmark datasets specifically designed for analyzing the… 
View on IEEE
arxiv.org
338 Citations
Highly Influential Citations
55
Background Citations
191
Methods Citations
153
Results Citations
9

338 Citations

Citation Type

Citation Type

Long-Term Visual Localization Revisited
TLDR
Three publicly available datasets containing images captured under a wide variety of viewing conditions, but lacking camera pose information, are extended, making evaluation of the impact of various factors on 6DOF camera pose estimation accuracy possible and the performance of state-of-the-art localization approaches on these datasets discussed.
Towards Robust Visual Localization in Challenging Conditions
TLDR
In this thesis, four papers are included, which present work towards solving the problem of long-term visual localization, and ideas for how semantic information may be included to aid in the localization process.
InLoc: Indoor Visual Localization with Dense Matching and View Synthesis
TLDR
A new large-scale visual localization method targeted for indoor spaces that significantly outperforms current state-of-the-art indoor localization approaches on this new challenging data.
VIZARD: Reliable Visual Localization for Autonomous Vehicles in Urban Outdoor Environments
TLDR
VIZARD, a visual localization system for urban outdoor environments, is presented, which is able to achieve nearly 100% recall with a localization accuracy below 0.5m under varying outdoor appearance conditions, including at night-time.
InLoc: Indoor Visual Localization with Dense Matching and View Synthesis
TLDR
A new large-scale visual localization method targeted for indoor environments that significantly outperforms current state-of-the-art indoor localization approaches on this new challenging data.
Improving Image Description with Auxiliary Modality for Visual Localization in Challenging Conditions
TLDR
This work proposes a new approach for outdoor large scale image-based localization that can deal with challenging scenarios like cross-season, cross-weather and day/night localization, and shows through extensive evaluation that this method can improve localization performances, especially in challenging scenarios when the visual appearance of the scene has changed.
Beyond Controlled Environments: 3D Camera Re-Localization in Changing Indoor Scenes
TLDR
This paper adapts 3RScan - a recently introduced indoor RGB-D dataset designed for object instance re-localization - to create RIO10, a new long-term camera re- localization benchmark focused on indoor scenes and explores how state-of-the-art cameraRe-localizers perform according to these metrics.
Learning Scene Geometry for Visual Localization in Challenging Conditions
TLDR
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 and a new learned global image descriptor that can effectively benefit from scene geometry information during training is proposed.
Pose Correction for Highly Accurate Visual Localization in Large-scale Indoor Spaces
TLDR
A novel module, named pose correction, is proposed that enables re-estimation of the pose with local feature matching in a similar view by reorganizing the local features and enhances the accuracy of the initially estimated pose and assigns more reliable ranks.
Visual Localization Using Sparse Semantic 3D Map
TLDR
Experiments on the challenging long-term visual localization benchmark dataset demonstrate that the proposed visual localization algorithm has significant improvement compared with the state-of-the-arts.
...
...

References

SHOWING 1-10 OF 84 REFERENCES
InLoc: Indoor Visual Localization with Dense Matching and View Synthesis
TLDR
A new large-scale visual localization method targeted for indoor environments that significantly outperforms current state-of-the-art indoor localization approaches on this new challenging data.
A Dataset for Benchmarking Image-Based Localization
TLDR
This paper addresses the problem of whether state-of-the-art visual localization techniques can be applied to tasks with demanding accuracy requirements and proposes a novel dataset for benchmarking image-based localization.
Are Large-Scale 3D Models Really Necessary for Accurate Visual Localization?
TLDR
It is demonstrated experimentally that large-scale 3D models are not strictly necessary for accurate visual localization, and it is shown that combining image-based methods with local reconstructions results in a pose accuracy similar to the state-of-the-art structure- based methods.
Semantic Visual Localization
TLDR
This paper proposes a novel approach based on a joint 3D geometric and semantic understanding of the world, enabling it to succeed under conditions where previous approaches failed, and employs a novel generative model for descriptor learning.
Summary Maps for Lifelong Visual Localization
TLDR
This work presents a framework for lifelong localization and mapping designed to provide robust and metrically accurate online localization in these kinds of changing environments, and presents a number of summary policies for selecting useful features for localization from the multisession map.
LSD-SLAM: Large-Scale Direct Monocular SLAM
TLDR
A novel direct tracking method which operates on \(\mathfrak{sim}(3)\), thereby explicitly detecting scale-drift, and an elegant probabilistic solution to include the effect of noisy depth values into tracking are introduced.
SeqSLAM: Visual route-based navigation for sunny summer days and stormy winter nights
TLDR
A new approach to visual navigation under changing conditions dubbed SeqSLAM, which removes the need for global matching performance by the vision front-end - instead it must only pick the best match within any short sequence of images.
Image-Based Localization Using LSTMs for Structured Feature Correlation
TLDR
Experimental results show the proposed CNN+LSTM architecture for camera pose regression for indoor and outdoor scenes outperforms existing deep architectures, and can localize images in hard conditions, where classic SIFT-based methods fail.
City-Scale Localization for Cameras with Known Vertical Direction
TLDR
This work considers the problem of localizing a novel image in a large 3D model, given that the gravitational vector is known, and extends accurate approximations and fast polynomial solvers to camera pose estimation.
PoseNet: A Convolutional Network for Real-Time 6-DOF Camera Relocalization
TLDR
This work trains a convolutional neural network to regress the 6-DOF camera pose from a single RGB image in an end-to-end manner with no need of additional engineering or graph optimisation, demonstrating that convnets can be used to solve complicated out of image plane regression problems.
...
...