Benchmarking 6DOF Outdoor Visual Localization in Changing Conditions

Sattler, Torsten; Maddern, Will; Toft, Carl; Torii, Akihiko; Hammarstrand, Lars; Stenborg, Erik; Safari, Daniel; Okutomi, Masatoshi; Pollefeys, Marc; Šivic, Josef; Kahl, Fredrik; Pajdla, Tomáš

doi:10.48550/arxiv.1707.09092

Cited by 4 publications

(6 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The vision [13] and robotics [14] communities have witnessed the rise of accurate and efficient image-based localization techniques that can be complementary to GPS, which are prone to error due to multi-path effects. The techniques can be classified into regression-based methods and retrievalbased ones.…”

Section: Related Workmentioning

confidence: 99%

“…On the contrary, retrievalbased ones are usually slower and have a large memory requirement for storing images or its descriptors for the entire city of globe. However, the retrieval-based methods typically provide higher accuracy and robustness [13].…”

Section: Related Workmentioning

confidence: 99%

“…Recently, it was shown that LSTMs can be used to achieve accurate localization in challenging lighting conditions [44]. A survey of different state-of-the-art localization techniques is given in [14], and there has been releases of many newer datasets [13], [45]. The idea of dominant set clustering is powerful for localization tasks [46].…”

Section: A Featuresmentioning

confidence: 99%

See 2 more Smart Citations

VLASE: Vehicle Localization by Aggregating Semantic Edges

Chaturvedi

Feng

et al. 2018

2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

View full text Add to dashboard Cite

In this paper, we propose VLASE, a framework to use semantic edge features from images to achieve on-road localization. Semantic edge features denote edge contours that separate pairs of distinct objects such as building-sky, roadsidewalk, and building-ground. While prior work has shown promising results by utilizing the boundary between prominent classes such as sky and building using skylines, we generalize this approach to consider semantic edge features that arise from 19 different classes. Our localization algorithm is simple, yet very powerful. We extract semantic edge features using a recently introduced CASENet architecture and utilize VLAD framework to perform image retrieval. Our experiments show that we achieve improvement over some of the state-of-the-art localization algorithms such as SIFT-VLAD and its deep variant NetVLAD. We use ablation study to study the importance of different semantic classes, and show that our unified approach achieves better performance compared to individual prominent features such as skylines.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

VLASE: Vehicle Localization by Aggregating Semantic Edges

Chaturvedi

Feng

et al. 2018

2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

View full text Add to dashboard Cite

show abstract

“…However, dynamic objects like vehicles and pedestrians, modification of the scene, weather conditions or different illumination break this assumption and result in increased number of failure. • Poses predicted by learning are not as accurate as structure-based methods [23]. In addition to larger error, they generate many outliers very far from the actual camera poses, so that trajectories are not consistent over time (in particular for monocular methods).…”

Section: Introductionmentioning

confidence: 99%

CoordiNet: uncertainty-aware pose regressor for reliable vehicle localization

Moreau¹,

Piasco²,

Tsishkou³

et al. 2021

Preprint

View full text Add to dashboard Cite

In this paper, we investigate visual-based camera localization with neural networks for robotics and autonomous vehicles applications. Our solution is a CNN-based algorithm which predicts camera pose (3D translation and 3D rotation) directly from a single image. It also provides an uncertainty estimate of the pose. Pose and uncertainty are learned together with a single loss function. Furthermore, we propose a new fully convolutional architecture, named CoordiNet, designed to embed some of the scene geometry.Our framework outperforms comparable methods on the largest available benchmark, the Oxford RobotCar dataset, with an average error of 8 meters where previous best was 19 meters. We have also investigated the performance of our method on large scenes for real time (18 fps) vehicle localization. In this setup, structure-based methods require a large database, and we show that our proposal is a reliable alternative, achieving 29cm median error in a 1.9km loop in a busy urban area.

show abstract

“…For example, most methods are brittle in certain scenarios, such as varying lighting conditions (e.g. changing time of day), differ- ent weather conditions or seasons [40], repetitive structures, textureless objects, extremely large viewpoint changes, dynamic elements within the environment, and faulty sensor calibration [4]. Because these situations are common in real-world scenarios, robust applications of those systems are difficult.…”

Section: Introductionmentioning

confidence: 99%

Global Pose Estimation with an Attention-Based Recurrent Network

Parisotto

Chaplot

Zhang

et al. 2018

2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)

View full text Add to dashboard Cite

The ability for an agent to localize itself within an environment is crucial for many real-world applications. For unknown environments, Simultaneous Localization and Mapping (SLAM) enables incremental and concurrent building of and localizing within a map. We present a new, differentiable architecture, Neural Graph Optimizer, progressing towards a complete neural network solution for SLAM by designing a system composed of a local pose estimation model, a novel pose selection module, and a novel graph optimization process. The entire architecture is trained in an end-to-end fashion, enabling the network to automatically learn domain-specific features relevant to the visual odometry and avoid the involved process of feature engineering. We demonstrate the effectiveness of our system on a simulated 2D maze and the 3D ViZ-Doom environment.

show abstract

Benchmarking 6DOF Outdoor Visual Localization in Changing Conditions

Cited by 4 publications

References 0 publications

VLASE: Vehicle Localization by Aggregating Semantic Edges

VLASE: Vehicle Localization by Aggregating Semantic Edges

CoordiNet: uncertainty-aware pose regressor for reliable vehicle localization

Global Pose Estimation with an Attention-Based Recurrent Network

Contact Info

Product

Resources

About