-
EdgeXAR: A 6-DoF Camera Multi-target Interaction Framework for MAR with User-friendly Latency Compensation
Authors:
Wenxiao Zhang,
Sikun Lin,
Farshid Hassani Bijarbooneh,
Haofei Cheng,
Tristan Braud,
Pengyuan Zhou,
Lik-hang Lee,
Pan Hui
Abstract:
The computational capabilities of recent mobile devices enable the processing of natural features for Augmented Reality (AR), but the scalability is still limited by the devices' computation power and available resources. In this paper, we propose EdgeXAR, a mobile AR framework that utilizes the advantages of edge computing through task offloading to support flexible camera-based AR interaction. W…
▽ More
The computational capabilities of recent mobile devices enable the processing of natural features for Augmented Reality (AR), but the scalability is still limited by the devices' computation power and available resources. In this paper, we propose EdgeXAR, a mobile AR framework that utilizes the advantages of edge computing through task offloading to support flexible camera-based AR interaction. We propose a hybrid tracking system for mobile devices that provides lightweight tracking with 6 Degrees of Freedom and hides the offloading latency from users' perception. A practical, reliable and unreliable communication mechanism is used to achieve fast response and consistency of crucial information. We also propose a multi-object image retrieval pipeline that executes fast and accurate image recognition tasks on the cloud and edge servers. Extensive experiments are carried out to evaluate the performance of EdgeXAR by building mobile AR Apps upon it. Regarding the Quality of Experience (QoE), the mobile AR Apps powered by EdgeXAR framework run on average at the speed of 30 frames per second with precise tracking of only 1~2 pixel errors and accurate image recognition of at least 97% accuracy. As compared to Vuforia, one of the leading commercial AR frameworks, EdgeXAR transmits 87% less data while providing a stable 30 FPS performance and reducing the offloading latency by 50 to 70% depending on the transmission medium. Our work facilitates the large-scale deployment of AR as the next generation of ubiquitous interfaces.
△ Less
Submitted 9 November, 2021;
originally announced November 2021.
-
CloudAR: A Cloud-based Framework for Mobile Augmented Reality
Authors:
Wenxiao Zhang,
Sikun Lin,
Farshid Hassani Bijarbooneh,
Hao Fei Cheng,
And Pan Hui
Abstract:
Computation capabilities of recent mobile devices enable natural feature processing for Augmented Reality (AR). However, mobile AR applications are still faced with scalability and performance challenges. In this paper, we propose CloudAR, a mobile AR framework utilizing the advantages of cloud and edge computing through recognition task offloading. We explore the design space of cloud-based AR ex…
▽ More
Computation capabilities of recent mobile devices enable natural feature processing for Augmented Reality (AR). However, mobile AR applications are still faced with scalability and performance challenges. In this paper, we propose CloudAR, a mobile AR framework utilizing the advantages of cloud and edge computing through recognition task offloading. We explore the design space of cloud-based AR exhaustively and optimize the offloading pipeline to minimize the time and energy consumption. We design an innovative tracking system for mobile devices which provides lightweight tracking in 6 degree of freedom (6DoF) and hides the offloading latency from users' perception. We also design a multi-object image retrieval pipeline that executes fast and accurate image recognition tasks on servers. In our evaluations, the mobile AR application built with the CloudAR framework runs at 30 frames per second (FPS) on average with precise tracking of only 1~2 pixel errors and image recognition of at least 97% accuracy. Our results also show that CloudAR outperforms one of the leading commercial AR framework in several performance metrics.
△ Less
Submitted 8 May, 2018;
originally announced May 2018.
-
Multitarget search on complex networks: A logarithmic growth of global mean random cover time
Authors:
Tongfeng Weng,
Jie Zhang,
Michael Small,
Ji Yang,
Farshid Hassani Bijarbooneh,
Pan Hui
Abstract:
We investigate multitarget search on complex networks and derive an exact expression for the mean random cover time that quantifies the expected time a walker needs to visit multiple targets. Based on this, we recover and extend some interesting results of multitarget search on networks. Specifically, we observe the logarithmic increase of the global mean random cover time with the target number f…
▽ More
We investigate multitarget search on complex networks and derive an exact expression for the mean random cover time that quantifies the expected time a walker needs to visit multiple targets. Based on this, we recover and extend some interesting results of multitarget search on networks. Specifically, we observe the logarithmic increase of the global mean random cover time with the target number for a broad range of random search processes, including generic random walks, biased random walks, and maximal entropy random walks. We show that the logarithmic growth pattern is a universal feature of multi-target search on networks by using the annealed network approach and the Sherman-Morrison formula. Moreover, we find that for biased random walks, the global mean random cover time can be minimized, and that the corresponding optimal parameter also minimizes the global mean first passage time, pointing towards its robustness. Our findings further confirm that the logarithmic growth pattern is a universal law governing multitarget search in confined media.
△ Less
Submitted 12 September, 2017; v1 submitted 12 January, 2017;
originally announced January 2017.
-
Dynamic Demand-Capacity Balancing for Air Traffic Management Using Constraint-Based Local Search: First Results
Authors:
Farshid Hassani Bijarbooneh,
Pierre Flener,
Justin Pearson
Abstract:
Using constraint-based local search, we effectively model and efficiently solve the problem of balancing the traffic demands on portions of the European airspace while ensuring that their capacity constraints are satisfied. The traffic demand of a portion of airspace is the hourly number of flights planned to enter it, and its capacity is the upper bound on this number under which air-traffic co…
▽ More
Using constraint-based local search, we effectively model and efficiently solve the problem of balancing the traffic demands on portions of the European airspace while ensuring that their capacity constraints are satisfied. The traffic demand of a portion of airspace is the hourly number of flights planned to enter it, and its capacity is the upper bound on this number under which air-traffic controllers can work. Currently, the only form of demand-capacity balancing we allow is ground holding, that is the changing of the take-off times of not yet airborne flights. Experiments with projected European flight plans of the year 2030 show that already this first form of demand-capacity balancing is feasible without incurring too much total delay and that it can lead to a significantly better demand-capacity balance.
△ Less
Submitted 7 October, 2009;
originally announced October 2009.
