A review of Visual-Based Localization

Xing, Xin; Jiang, Jie; Zou, Yin

doi:10.1145/3366194.3366211

Cited by 16 publications

(8 citation statements)

References 71 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The numerous applications of camera pose estimation and image localization render it as an interesting and active field of research. Traditionally, there are two distinct approaches to estimate the position and orientation of a camera [Xin et al, 2019]: (i) Features based methods and (ii) Network based pose regression methods. The proposed method forms a new category of possible approaches: (iii) Direct 2D-3D descriptor matching based methods.…”

Section: Related Workmentioning

confidence: 99%

“…The regression networks based methods (e.g. Kendall et al [2015]; Kendall and Cipolla [2017]; Walch et al [2017]), use deep neural networks to es-timate the pose of the camera and consequently have high requirements for computational resources such as powerful GPUs and require a lot of training data [Xin et al, 2019] from different viewpoints to ensure that the poses of query cameras are sufficiently close to the training ones [Sattler et al, 2019]. Moreover they scale poorly with the increase in the size of 3D models and usually run into the problems of nonconvergence for end-to-end training [Brachmann and Rother, 2018].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Unconstrained Matching of 2D and 3D Descriptors for 6-DOF Pose Estimation

Nadeem,

Bennamoun,

Togneri

et al. 2020

Preprint

View full text Add to dashboard Cite

This paper proposes a novel concept to directly match feature descriptors extracted from 2D images with feature descriptors extracted from 3D point clouds. We use this concept to directly localize images in a 3D point cloud. We generate a dataset of matching 2D and 3D points and their corresponding feature descriptors, which is used to learn a Descriptor-Matcher classifier. To localize the pose of an image at test time, we extract keypoints and feature descriptors from the query image. The trained Descriptor-Matcher is then used to match the features from the image and the point cloud. The locations of the matched features are used in a robust pose estimation algorithm to predict the location and orientation of the query image. We carried out an extensive evaluation of the proposed method for indoor and outdoor scenarios and with different types of point clouds to verify the feasibility of our approach. Experimental results demonstrate that direct matching of feature descriptors from images and point clouds is not only a viable idea but can also be reliably used to estimate the 6-DOF poses of query cameras in any type of 3D point cloud in an unconstrained manner with high precision.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Unconstrained Matching of 2D and 3D Descriptors for 6-DOF Pose Estimation

Nadeem,

Bennamoun,

Togneri

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

“…A growing number of autonomous transportation systems are being developed for safety-critical urban applications [1], including traffic management, lane keeping, ride-share services, cargo transport, and medicalaid delivery. With recent advances in sensor technologies and computing resources, there is an increasing interest in using Artificial Intelligence (AI)-driven tools for localization of autonomous vehicles, such as selfdriving cars and unmanned aerial vehicles (UAVs) [2,3,4]. Specifically, AI tools are popular in localization constellations, such as BeiDou and Galileo, and multiple frequencies, such as the GPS L1 and L5 signals [23].…”

Section: Introductionmentioning

confidence: 99%

Mosaic Zonotope Shadow Matching for Risk-Aware Autonomous Localization in Harsh Urban Environments

Neamati¹,

Bhamidipati²,

Gao³

2022

Preprint

View full text Add to dashboard Cite

Risk-aware urban localization with the Global Navigation Satellite System (GNSS) remains an unsolved problem with frequent misdetection of the user's street or side of the street. Significant advances in 3D mapaided GNSS use grid-based GNSS shadow matching alongside AI-driven line-of-sight (LOS) classifiers and server-based processing to improve localization accuracy, especially in the cross-street direction. Our prior work introduces a new paradigm for shadow matching that proposes set-valued localization with computationally efficient zonotope set representations. While existing literature improved accuracy and efficiency, the current state of shadow matching theory does not address the needs of risk-aware autonomous systems. We extend our prior work to propose Mosaic Zonotope Shadow Matching (MZSM) that employs a classifier-agnostic polytope mosaic architecture to provide risk-awareness and certifiable guarantees on urban positioning. We formulate a recursively expanding binary tree that refines an initial location estimate with set operations into smaller polytopes. Together, the smaller polytopes form a mosaic. We weight the tree branches with the probability that the user is in line of sight of the satellite and expand the tree with each new satellite observation. Our method yields an exact shadow matching distribution from which we guarantee uncertainty bounds on the user localization. We perform high-fidelity simulations using a 3D building map of San Francisco to validate our algorithm's risk-aware improvements. We demonstrate that MZSM provides certifiable guarantees across varied data-driven LOS classifier accuracies and yields a more precise understanding of the uncertainty over existing methods. We validate that our tree-based construction is efficient and tractable, computing a mosaic from 14 satellites in 0.63 seconds and growing quadratically in the satellite number.

show abstract

“…Visual geo-location is an important branch of computer vision [1][2][3]. By extracting visual features from images and comparing with the corresponding datasets, the position and the orientation (i.e., heading angle, pitch and roll) of the camera used can be obtained [4,5].…”

Section: Introductionmentioning

confidence: 99%

Fast Geo-Location Method Based on Panoramic Skyline in Hilly Area

Pan

Tang

Tjahjadi

et al. 2021

IJGI

View full text Add to dashboard Cite

Localization method based on skyline for visual geo-location is an important auxiliary localization method that does not use a satellite positioning system. Due to the computational complexity, existing panoramic skyline localization methods determine a small area using prior knowledge or auxiliary sensors. After correcting the camera orientation using inertial navigation sensors, a fine position is achieved via the skyline. In this paper, a new panoramic skyline localization method is proposed that involves the following. By clustering the sampling points in the location area and improving the existing retrieval method, the computing efficiency of the panoramic skyline localization is increased by fourfold. Furthermore, the camera orientation is estimated accurately from the terrain features in the image. Experimental results show that the proposed method achieves higher localization accuracy and requires less computation for a large area without the aid of external sensors.

show abstract

A review of Visual-Based Localization

Cited by 16 publications

References 71 publications

Unconstrained Matching of 2D and 3D Descriptors for 6-DOF Pose Estimation

Unconstrained Matching of 2D and 3D Descriptors for 6-DOF Pose Estimation

Mosaic Zonotope Shadow Matching for Risk-Aware Autonomous Localization in Harsh Urban Environments

Fast Geo-Location Method Based on Panoramic Skyline in Hilly Area

Contact Info

Product

Resources

About