Real-time indoor assistive localization with mobile omnidirectional vision and cloud GPU acceleration

Feng Hu; Zhigang Zhu; Jeury Mejia; Hao Tang; Jianting Zhang; Feng Hu; Zhigang Zhu; Jeury Mejia; Hao Tang; Jianting Zhang

doi:10.3934/ElectrEng.2017.1.74

AIMS Electronics and Electrical Engineering

2017, Volume 1, Issue 1: 74-99. doi: 10.3934/ElectrEng.2017.1.74

Previous Article Next Article

Research article

Real-time indoor assistive localization with mobile omnidirectional vision and cloud GPU acceleration

1.
Department of Computer Science, The Graduate Center, City University of New York, New York City, NY, United States
2.
Department of Computer Science, The City College, City University of New York, New York City, NY, United States
3.
Department of Computer Science, Rutgers University, New Brunswick, NJ, United States
4.
Department of Computer Information Systems, Borough of Manhattan Community College, City University of New York, New York City, NY, United States

Received: 15 August 2017 Accepted: 28 November 2017 Published: 13 December 2017

In this paper we propose a real-time assistive localization approach to help blind and visually impaired people in navigating an indoor environment. The system consists of a mobile vision front end with a portable panoramic lens mounted on a smart phone, and a remote image feature-based database of the scene on a GPU-enabled server. Compact and e ective omnidirectional image features are extracted and represented in the smart phone front end, and then transmitted to the server in the cloud. These features of a short video clip are used to search the database of the indoor environment via image-based indexing to find the location of the current view within the database, which is associated with floor plans of the environment. A median-filter-based multi-frame aggregation strategy is used for single path modeling, and a 2D multi-frame aggregation strategy based on the candidates’ distribution densities is used for multi-path environmental modeling to provide a final location estimation. To deal with the high computational cost in searching a large database for a realistic navigation application, data parallelism and task parallelism properties are identified in the database indexing process, and computation is accelerated by using multi-core CPUs and GPUs. User-friendly HCI particularly for the visually impaired is designed and implemented on an iPhone, which also supports system configurations and scene modeling for new environments. Experiments on a database of an eight-floor building are carried out to demonstrate the capacity of the proposed system, with real-time response (14 fps) and robust localization results.
- assistive indoor localization,
- real-time system,
- GPU acceleration,
- mobile computing,
- omnidirectional vision
Citation: Feng Hu, Zhigang Zhu, Jeury Mejia, Hao Tang, Jianting Zhang. Real-time indoor assistive localization with mobile omnidirectional vision and cloud GPU acceleration[J]. AIMS Electronics and Electrical Engineering, 2017, 1(1): 74-99. doi: 10.3934/ElectrEng.2017.1.74

Related Papers:

Abstract

In this paper we propose a real-time assistive localization approach to help blind and visually impaired people in navigating an indoor environment. The system consists of a mobile vision front end with a portable panoramic lens mounted on a smart phone, and a remote image feature-based database of the scene on a GPU-enabled server. Compact and e ective omnidirectional image features are extracted and represented in the smart phone front end, and then transmitted to the server in the cloud. These features of a short video clip are used to search the database of the indoor environment via image-based indexing to find the location of the current view within the database, which is associated with floor plans of the environment. A median-filter-based multi-frame aggregation strategy is used for single path modeling, and a 2D multi-frame aggregation strategy based on the candidates’ distribution densities is used for multi-path environmental modeling to provide a final location estimation. To deal with the high computational cost in searching a large database for a realistic navigation application, data parallelism and task parallelism properties are identified in the database indexing process, and computation is accelerated by using multi-core CPUs and GPUs. User-friendly HCI particularly for the visually impaired is designed and implemented on an iPhone, which also supports system configurations and scene modeling for new environments. Experiments on a database of an eight-floor building are carried out to demonstrate the capacity of the proposed system, with real-time response (14 fps) and robust localization results.

References

[1]	Cummins M, Newman P (2011) Appearance-only slam at large scale with fab-map 2.0. Int J Robot Res 30: 1100-1123. doi: 10.1177/0278364910385483
[2]	Davison AJ, Reid ID, Molton ND, et al. (2007) Monoslam: Real-time single camera slam. IEEE T Pattern Anal 29: 1052-1067. doi: 10.1109/TPAMI.2007.1049
[3]	Di Corato F, Pollini L, Innocenti M, et al. (2011) An entropy-like approach to vision based autonomous navigation. IEEE T Robotic Autom 1640-1645.
[4]	Rivera-Rubio J, Idrees S, Alexiou I, et al. (2013) Mobile visual assistive apps: Benchmarks of vision algorithm performance. In: New trends in image analysis and processing–iciap Springer 30-40.
[5]	Liu Z, Zhang L, Liu Q, et al. (2017) Fusion of magnetic and visual sensors for indoor localization: Infrastructure-free and more effective. IEEE T Multimedia 19: 874-888. doi: 10.1109/TMM.2016.2636750
[6]	Tang H, Tsering N, Hu F (2016) Automatic pre-journey indoor map generation using autocad floor plan. J Tec Pers Disabil 4: 176-191.
[7]	Ravi N, Shankar P, Frankel A, et al. (2006) Indoor localization using camera phones. In: IEEE Workshop on Mobile Computing Systems and Applications IEEE 49.
[8]	Tang H, Tsering N, Hu F (2016) Automatic pre-journey indoor map generation using autocad floor plan. In: 31st Annual International Technology and Persons with Disabilities Conference San Diego, CA, U.S.A.
[9]	Kulyukin V, Gharpure C, Nicholson J, et al. (2004) Rfid in robot-assisted indoor navigation for the visually impaired. Intell Robot Syst 2: 1979-1984.
[10]	Cicirelli G, Milella A, Di Paola D (2012) Rfid tag localization by using adaptive neuro-fuzzy inference for mobile robot applications. Ind Robot 39: 340-348. doi: 10.1108/01439911211227908
[11]	Legge GE, Beckmann PJ, Tjan BS, et al. (2013) Indoor navigation by people with visual impairment using a digital sign system. PloS one 8.
[12]	Hu F, Zhu Z, Zhang J (2014) Mobile panoramic vision for assisting the blind via indexing and localization. In: Computer Vision-ECCV 2014 Workshops Springer 600-614.
[13]	Murillo AC, Singh G, Kosecka J, et al. (2013) Localization in urban environments using a panoramic gist descriptor. IEEE T Robot 29: 146-160. doi: 10.1109/TRO.2012.2220211
[14]	Kume H, Suppé A, Kanade T (2013) Vehicle localization along a previously driven route using an image database. J Mach Vision Appl 177-180.
[15]	Badino H, Huber D, Kanade T (2012) Real-time topometric localization. IEEE Int Conference Robotic Autom (ICRA) 1635-1642.
[16]	Hu F, Tsering N, Tang H, et al. (2016) Indoor localization for the visually impaired using a 3d sensor. J Technol Pers Disabil 4: 192-203.
[17]	Lee YH, Medioni G (2014) Wearable rgbd indoor navigation system for the blind. In: Workshop on Assistive Computer Vision and Robotics IEEE 493-508.
[18]	Hu F, Tsering N, Tang H, et al. (2016) Rgb-d sensor based indoor localization for the visually impaired. In: 31st Annual International Technology and Persons with Disabilities Conference San Diego, CA, U.S.A.
[19]	Molina E, Zhu Z (2013) Visual noun navigation framework for the blind. J Assist Technol 7: 118-130. doi: 10.1108/17549451311328790
[20]	GoPano (2016) Gopano micro camera adapter. Available from: Available from: http://www.gopano.com/products/gopano-micro.
[21]	Hu F, Li T, Geng Z (2011) Constraints-based graph embedding optimal surveillance-video mosaicing. Asian Conf Pattern Recogn (ACPR) 311-315.
[22]	Aly M, Bouguet JY (2012) Street view goes indoors: Automatic pose estimation from uncalibrated unordered spherical panoramas. Workshop Appl Comput Vision (WACV) 1-8.
[23]	Oliva A, Torralba A (2001) Modeling the shape of the scene: A holistic representation of the spatial envelope. Int J Comput Vision 42: 145-175. doi: 10.1023/A:1011139631724
[24]	Farinella G, Battiato S (2011) Scene classification in compressed and constrained domain. IET Comput Vision 5: 320-334. doi: 10.1049/iet-cvi.2010.0056
[25]	Xiao J, Ehinger KA, Oliva A, et al. (2012) Recognizing scene viewpoint using panoramic place representation. Comput Vision Pattern Recogn (CVPR) 2695-2702.
[26]	Zhu Z, Yang S, Xu G, et al. (1998) Fast road classification and orientation estimation using omniview images and neural networks. IEEE T Image Process 7: 1182-1197. doi: 10.1109/83.704310
[27]	Sun W, Duan N, Ji P, et al. (2016) Intelligent in-vehicle air quality management: A smart mobility application dealing with air pollution in the traffc. In: 23rd World Congress on Intelligent Transportation Systems Melbourne Australia.
[28]	Ma C, Duan N, SunW, et al. (2017) Reducing air pollution exposure in a road trip. In: 24rd World Congress on Intelligent Transportation Systems Montreal, Canada.
[29]	Farinella G, Rav D, Tomaselli V, et al. (2015) Representing scenes for real-time context classification on mobile devices. Pattern Recogn 48: 1086-1100. doi: 10.1016/j.patcog.2014.05.014
[30]	Altwaijry H, Moghimi M, Belongie S (2014) Recognizing locations with google glass: A case study. IEEE Winter Conf Appl Comput Vision (WACV) Colorado.
[31]	Paisios N (2012) Mobile accessibility tools for the visually impaired [dissertation]. New York University.
[32]	Manduchi R (2012) Mobile vision as assistive technology for the blind: An experimental study. In: The 13th International Conference on Computers Helping People of Special Needs (ICCHP) Springer.
[33]	Scaramuzza D, Martinelli A, Siegwart R (2006) A toolbox for easily calibrating omnidirectional cameras. Intell Robot Syst 5695-5701.
[34]	Nayar SK (1997) Catadioptric omnidirectional camera. Comput Vision Pattern Recogn 482-488.
[35]	Kang SB (2000) Catadioptric self-calibration. Comput Vision Pattern Recogn 1: 201-207.
[36]	Scaramuzza D, Martinelli A, Siegwart R (2006) A flexible technique for accurate omnidirectional camera calibration and structure from motion. Int Conf Comput Vision Syst 45-45.
[37]	Zhu Z, Rajasekar KD, Riseman EM, et al. (2000) Panoramic virtual stereo vision of cooperative mobile robots for localizing 3d moving objects. Omnidirectional Vision 29-36.

Reader Comments

Your name:*

Email:*
© 2017 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)