Show simple item record

dc.contributor.authorYaseen, Muhammad Usman
dc.contributor.authorAnjum, Ashiq
dc.contributor.authorAntonopoulos, Nikolaos
dc.date.accessioned2017-12-20T16:37:29Z
dc.date.available2017-12-20T16:37:29Z
dc.date.issued2017-12-05
dc.identifier.citationYaseen, M. U. et al (2017) 'Modeling and analysis of a deep learning pipeline for cloud based video analytics.', Proceedings of the Fourth IEEE/ACM International Conference on Big Data Computing, Applications and Technologies (BDCAT 2017), pp. 121-130.en
dc.identifier.isbn9781450355490
dc.identifier.doi10.1145/3148055.3148081
dc.identifier.urihttp://hdl.handle.net/10545/622032
dc.description.abstractVideo analytics systems based on deep learning approaches are becoming the basis of many widespread applications including smart cities to aid people and traffic monitoring. These systems necessitate massive amounts of labeled data and training time to perform fine tuning of hyper-parameters for object classification. We propose a cloud based video analytics system built upon an optimally tuned deep learning model to classify objects from video streams. The tuning of the hyper-parameters including learning rate, momentum, activation function and optimization algorithm is optimized through a mathematical model for efficient analysis of video streams. The system is capable of enhancing its own training data by performing transformations including rotation, flip and skew on the input dataset making it more robust and self-adaptive. The use of in-memory distributed training mechanism rapidly incorporates large number of distinguishing features from the training dataset - enabling the system to perform object classification with least human assistance and external support. The validation of the system is performed by means of an object classification case-study using a dataset of 100GB in size comprising of 88,432 video frames on an 8 node cloud. The extensive experimentation reveals an accuracy and precision of 0.97 and 0.96 respectively after a training of 6.8 hours. The system is scalable, robust to classification errors and can be customized for any real-life situation.
dc.description.sponsorshipN/Aen
dc.language.isoenen
dc.relation.urlhttp://dl.acm.org/citation.cfm?doid=3148055.3148081en
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/*
dc.subjectVideo analysisen
dc.subjectDeep learningen
dc.subjectCloud computingen
dc.titleModeling and analysis of a deep learning pipeline for cloud based video analytics.en
dc.typeMeetings and Proceedingsen
dc.contributor.departmentUniversity of Derbyen
dc.identifier.journalProceedings of the Fourth IEEE/ACM International Conference on Big Data Computing, Applications and Technologies (BDCAT 2017)en
dc.contributor.institutionUniversity of Derby, Derby, United Kingdom
dc.contributor.institutionUniversity of Derby, Derby, United Kingdom
dc.contributor.institutionUniversity of Derby, Derby, United Kingdom
refterms.dateFOA2019-02-28T16:24:25Z
html.description.abstractVideo analytics systems based on deep learning approaches are becoming the basis of many widespread applications including smart cities to aid people and traffic monitoring. These systems necessitate massive amounts of labeled data and training time to perform fine tuning of hyper-parameters for object classification. We propose a cloud based video analytics system built upon an optimally tuned deep learning model to classify objects from video streams. The tuning of the hyper-parameters including learning rate, momentum, activation function and optimization algorithm is optimized through a mathematical model for efficient analysis of video streams. The system is capable of enhancing its own training data by performing transformations including rotation, flip and skew on the input dataset making it more robust and self-adaptive. The use of in-memory distributed training mechanism rapidly incorporates large number of distinguishing features from the training dataset - enabling the system to perform object classification with least human assistance and external support. The validation of the system is performed by means of an object classification case-study using a dataset of 100GB in size comprising of 88,432 video frames on an 8 node cloud. The extensive experimentation reveals an accuracy and precision of 0.97 and 0.96 respectively after a training of 6.8 hours. The system is scalable, robust to classification errors and can be customized for any real-life situation.


Files in this item

Thumbnail
Name:
p121-yaseen.pdf
Size:
2.268Mb
Format:
PDF
Description:
Author Accepted Manuscript

This item appears in the following Collection(s)

Show simple item record

http://creativecommons.org/licenses/by/4.0/
Except where otherwise noted, this item's license is described as http://creativecommons.org/licenses/by/4.0/