Multiclass object recognition using smart phone and cloud computing for augmented reality and video surveillance applications

Paul, Anjan Kumar; Park, Jong Sou

doi:10.1109/iciev.2013.6572719

Cited by 8 publications

(3 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…fr bw fr f f fr re cm B L R (7) PSNR constraint The video PSNR should be greater than or equal to the required minimum PSNR, i.e., _ _ m i n ( ) sn sn f cm R (8) Client capability constraint The required processing capability for decoding video at client side should be less than or equal to what client device owns, i.e., _ ( , , ) de f fr re cm P (9) Other constraints All variables should be greater or equal to 0, i.e., , , , , 0 i n re fr cm g i (10) If we consider the encoding task is handled by one VM, the following constraint should be included as well:…”

Section: Resolution Constraintmentioning

confidence: 99%

“…Real-time face recognition approach is implemented using a mobile-cloudlet-cloud acceleration architecture in [9]. In another study [8], multiclass object recognition using smart phone and cloud computing for augmented reality and video surveillance applications is proposed.…”

Section: Related Workmentioning

confidence: 99%

“…Assume that the quality of video is guaranteed by constraints of (5), (7) and (8), we remove the corresponding quality factor from the optimization goal. The non-linear programming model of the M-JPEG encoder can be re-designed as follows: min We directly use (13), (14), (18) and (19) to replace (4), (5), (8) and (9), respectively. The constraint (6) is substituted by (15) and (16) to guarantee the provided CPU and GPU capacity is sufficient enough to meet the computational resource requirement of encoding setting j .…”

Section: Function Specificationmentioning

confidence: 99%

See 2 more Smart Citations

Novel remote display method for multimedia cloud

Song

Tian

Hassan

et al. 2014

2014 IEEE International Conference on Multimedia and Expo Workshops (ICMEW)

View full text Add to dashboard Cite

Cloud computing offers sufficient computing and storage resources that can be used to provide multimedia services. Migrating the existing multimedia service to cloud brings a new challenging issue, i.e., remote display of video contents. To reduce the bandwidth consumption especially for mobile users, it is desired to encode video before sending to client. Existing encoding methods have unique advantages and disadvantages, differing their performance under varying situations. Thus, we propose to use multiencoder method to solve the remote display problem for remote multimedia cloud. To select the most appropriate encoder, factors including cost, application requirement, network, client device and codec implementation are considered. In this paper, we form a non-linear programming model, and take M-JPEG encoder as an example to illustrate how to apply the proposed model for getting desired optimization.

show abstract

Section: Resolution Constraintmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%