Scalable Place Recognition Under Appearance Change for Autonomous Driving

Doan, Anh-Dzung; Latif, Yasir; Chin, Tat-Jun; Liu, Yu; Do, Thanh-Toan; Reid, Ian

doi:10.1109/iccv.2019.00941

Cited by 72 publications

(39 citation statements)

References 39 publications

(67 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Key to this system is a novel, principled technique for conditioning transition probabilities between discrete states with continuous 3-dof odometry. This extends existing discrete filters which use none [10], [17] to limited [9] odometry and provides an alternative to continuous state systems utilizing particle filters [10], [18].…”

Section: Introductionmentioning

confidence: 95%

Probabilistic Appearance-Invariant Topometric Localization With New Place Awareness

Mei

Fischer

Sünderhauf

et al. 2021

IEEE Robot. Autom. Lett.

View full text Add to dashboard Cite

Probabilistic state-estimation approaches offer a principled foundation for designing localization systems, because they naturally integrate sequences of imperfect motion and exteroceptive sensor data. Recently, probabilistic localization systems utilizing appearance-invariant visual place recognition (VPR) methods as the primary exteroceptive sensor have demonstrated state-of-the-art performance in the presence of substantial appearance change. However, existing systems 1) do not fully utilize odometry data within the motion models, and 2) are unable to handle route deviations, due to the assumption that query traverses exactly repeat the mapping traverse. To address these shortcomings, we present a new probabilistic topometric localization system which incorporates full 3-dof odometry into the motion model and furthermore, adds an "off-map" state within the state-estimation framework, allowing query traverses which feature significant route detours from the reference map to be successfully localized. We perform extensive evaluation on multiple query traverses from the Oxford RobotCar dataset exhibiting both significant appearance change and deviations from routes previously traversed. In particular, we evaluate performance on two practically relevant localization tasks: loop closure detection and global localization. Our approach achieves major performance improvements over both existing and improved state-of-the-art systems.

show abstract

Section: Introductionmentioning

confidence: 95%

Probabilistic Appearance-Invariant Topometric Localization With New Place Awareness

Mei

Fischer

Sünderhauf

et al. 2021

IEEE Robot. Autom. Lett.

View full text Add to dashboard Cite

show abstract

“…Disregarding the collection costs, this idea is difficult to implement as the dimension of the database, and consequently the computational cost of the place recognition algorithm, can quickly become unmanageable. To solve this scalability problem, Doan et al [196] propose a solution that consists of three elements. Firstly, a VPR algorithm based on a Hidden Markov Model that is efficient both in terms of training time and testing time.…”

Section: E Adapting To Different Environmental Conditionsmentioning

confidence: 99%

“…Metric information can also be used to combine newly created graphs with previous acquisition, by merging close vertices into a single location [182]. Doan et al [196] have recently proposed a strategy to efficiently expand a map. This strategy is based on two different optimizations: i) adding images that provide only new information based on the localization belief (culling); ii) merging nodes that refer to the same place but were visited in different occasions (compression).…”

Section: A Mapsmentioning

confidence: 99%

“…Although FAB-MAP can work as a pure image retrieval process by always assuming a uniform location prior, using even a simple motion model to leverage the prior estimate improves performance. Hidden Markov Models (HMM) are also used extensively for visual place recognition in robotics to exploit the temporal order of the captured images and the high correlation between time and place due to motion constraints [193], [196]. In [193], a HMM is used to combine image representations produced by four methods: two hand-crafted representations and two types of deep learned representations.…”

Section: ) Stochastic Modelsmentioning

confidence: 99%

See 1 more Smart Citation

A Survey on Deep Visual Place Recognition

2021

View full text Add to dashboard Cite

In recent years visual place recognition (VPR), i.e., the problem of recognizing the location of images, has received considerable attention from multiple research communities, spanning from computer vision to robotics and even machine learning. This interest is fueled on one hand by the relevance that visual place recognition holds for many applications and on the other hand by the unsolved challenge of making these methods perform reliably in different conditions and environments. This paper presents a survey of the state-of-the-art of research on visual place recognition, focusing on how it has been shaped by the recent advances in deep learning. We start discussing the image representations used in this task and how they have evolved from using hand-crafted to deep-learned features. We further review how metric learning techniques are used to get more discriminative representations, as well as techniques for dealing with occlusions, distractors, and shifts in the visual domain of the images. The survey also provides an overview of the specific solutions that have been proposed for applications in robotics and with aerial imagery. Finally the survey provides a summary of datasets that are used in visual place recognition, highlighting their different characteristics. INDEX TERMS Visual place recognition, image representation learning, deep learning.

show abstract

“…Localization: Over the last two decades, the robotics community extensively considered the problem of localization and mapping [15] involving a diverse set of sensors, most prominently cameras [16] and lidars [2]- [4]. Particularly for autonomous driving, a considerable amount of work focused on dealing with challenging and changing appearance conditions [9] such as weather [17] or occlusions [7], [18]. To improve robustness, radar has also been considered as a localization modality for autonomous driving [19]- [22].…”

Section: Related Workmentioning

confidence: 99%

GROUNDED: The Localizing Ground Penetrating Radar Evaluation Dataset

Ort¹,

Gilitschenski²,

Rus³

2021

Robotics: Science and Systems XVII

View full text Add to dashboard Cite

Mapping and localization using surface features is prone to failure due to environment changes such as inclement weather. Recently, Localizing Ground Penetrating Radar (LGPR) has been proposed as an alternative means of localizing using underground features that are stable over time and less affected by surface conditions. However, due to the lack of commercially available LGPR sensors, the wider research community has been largely unable to replicate this work or build new and innovative solutions. We present GROUNDED an open dataset of LGPR scans collected in a variety of environments and weather conditions. By labeling this data with ground truth localization from an RTK-GPS / Inertial Navigation System, and carefully calibrating and time synchronizing the radar scans with ground truth positions, camera imagery, and Lidar data, we enable researchers to build novel localization solutions that are resilient to changing surface conditions. We include 108 individual runs totalling 450 km of driving with LGPR, GPS, Odometry, Camera, and Lidar measurements. We also present two new evaluation benchmarks for 1) Localizing in Weather and 2) Multi-lane Mapping, to enable comparisons of future work supported by the dataset. The dataset can be accessed at http://lgprdata.com.

show abstract

Scalable Place Recognition Under Appearance Change for Autonomous Driving

Cited by 72 publications

References 39 publications

Probabilistic Appearance-Invariant Topometric Localization With New Place Awareness

Probabilistic Appearance-Invariant Topometric Localization With New Place Awareness

A Survey on Deep Visual Place Recognition

GROUNDED: The Localizing Ground Penetrating Radar Evaluation Dataset

Contact Info

Product

Resources

About