Real-Time Visual Place Recognition for Personal Localization on a Mobile Device

Nowicki, Michał; Wietrzykowski, Jan; Skrzypczyński, Piotr

doi:10.1007/s11277-017-4502-y

Cited by 12 publications

(10 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In contrast, our paper [6] focused on determining additional localization constraints from qualitative visual observations and direct user input, but only a brief feasibility study was presented without thorough evaluation. In this article, we extend the previous research by completely reworking the VPR, as it no longer requires user input and is based on sequences of images, like in [35].…”

Section: Related Workmentioning

confidence: 91%

“…This idea was extended in [34] utilizing a single global Local Difference Binary (LDB) descriptor per image. We have shown in our previous work [35] that this method can be implemented in a smartphone, presenting FastABLE—a significantly faster version of the ABLE-M algorithm from [34].…”

Section: Related Workmentioning

confidence: 99%

“…In this article, we propose to directly focus on comparing sequences of recorded images that are greatly reduced to capture only the most important aspects of the image. The proposed solution is based on the FastABLE [35] algorithm that is a faster modification of the original ABLE-M/OpenABLE solution [44]. As all these VPR systems informed only about the visual similarity, we introduce a procedure that treats the matching images similarly as the matching WiFi scans are used in the WKNN algorithm.…”

Section: Vprmentioning

confidence: 99%

“…As a result, it is possible to greatly reduce the size of individual images providing the VPR system with additional robustness to image acquisition problems or small changes in the environment. Drastic changes in the environment or crowds in front of the camera might prevent the system from correctly recognizing the place but even in these conditions satisfactory results were reported as shown in [35].…”

Section: Vprmentioning

confidence: 99%

See 3 more Smart Citations

Leveraging Visual Place Recognition to Improve Indoor Positioning with Limited Availability of WiFi Scans

Nowicki

Skrzypczyński

2019

Sensors

Self Cite

View full text Add to dashboard Cite

WiFi-based fingerprinting is promising for practical indoor localization with smartphones because this technique provides absolute estimates of the current position, while the WiFi infrastructure is ubiquitous in the majority of indoor environments. However, the application of WiFi fingerprinting for positioning requires pre-surveyed signal maps and is getting more restricted in the recent generation of smartphones due to changes in security policies. Therefore, we sought new sources of information that can be fused into the existing indoor positioning framework, helping users to pinpoint their position, even with a relatively low-quality, sparse WiFi signal map. In this paper, we demonstrate that such information can be derived from the recognition of camera images. We present a way of transforming qualitative information of image similarity into quantitative constraints that are then fused into the graph-based optimization framework for positioning together with typical pedestrian dead reckoning (PDR) and WiFi fingerprinting constraints. Performance of the improved indoor positioning system is evaluated on different user trajectories logged inside an office building at our University campus. The results demonstrate that introducing additional sensing modality into the positioning system makes it possible to increase accuracy and simultaneously reduce the dependence on the quality of the pre-surveyed WiFi map and the WiFi measurements at run-time.

show abstract

Section: Related Workmentioning

confidence: 91%

Section: Related Workmentioning

confidence: 99%

Section: Vprmentioning

confidence: 99%

Section: Vprmentioning

confidence: 99%

See 2 more Smart Citations

Leveraging Visual Place Recognition to Improve Indoor Positioning with Limited Availability of WiFi Scans

Nowicki

Skrzypczyński

2019

Sensors

Self Cite

View full text Add to dashboard Cite

show abstract

“…Systems utilizing camera images [4,5], Bluetooth Low Energy [6], anomalies of the ambient magnetic field [7], and even light sources [8] have been proposed for localization with smartphones. A brief survey of the indoor positioning techniques focusing on the suitability of different sensing modalities is presented in Ref.…”

Section: Introductionmentioning

confidence: 99%

A Multi-User Personal Indoor Localization System Employing Graph-Based Optimization

Nowicki

Skrzypczyński

2019

Sensors

Self Cite

View full text Add to dashboard Cite

Personal indoor localization with smartphones is a well-researched area, with a number of approaches solving the problem separately for individual users. Most commonly, a particle filter is used to fuse information from dead reckoning and WiFi or Bluetooth adapters to provide an accurate location of the person holding a smartphone. Unfortunately, the existing solutions largely ignore the gains that emerge when a single localization system estimates locations of multiple users in the same environment. Approaches based on filtration maintain only estimates of the current poses of the users, marginalizing the historical data. Therefore, it is difficult to fuse data from multiple individual trajectories that are usually not perfectly synchronized in time. We propose a system that fuses the information from WiFi and dead reckoning employing the graph-based optimization, which is widely applied in robotics. The presented system can be used for localization of a single user, but the improvement is especially visible when this approach is extended to a multi-user scenario. The article presents a number of experiments performed with a smartphone inside an office building. These experiments demonstrate that graph-based optimization can be used as an efficient fusion mechanism to obtain accurate trajectory estimates both in the case of a single user and in a multi-user indoor localization system. The code of our system together with recorded dataset will be made available when the paper gets published.

show abstract