Squeezing Deep Learning Into Mobile Devices
Ramya Venkatesh1, Ramesh Ragala2, R. Jagadeesh Kannan3
1Ramya Venkatesh, M.Tech, SCSE, VIT University, Chennai Campus (Tamil Nadu), India.
2Ramesh Ragala, Assistant Professor, SCSE, VIT University, Chennai Campus (Tamil Nadu), India.
3R.Jagadeesh Kannan, Professor, SCSE, VIT University, Chennai Campus (Tamil Nadu), India.
Manuscript received on 27 March 2019 | Revised Manuscript received on 08 April 2019 | Manuscript Published on 18 April 2019 | PP: 902-906 | Volume-7 Issue-6S March 2019 | Retrieval Number: F03840376S19/2019©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Advancement in technology has led to generation of data in huge amount every second. Most of the data is generated by mobile devices. Various sensors in mobile devices continuously keep generating the data. This data can be used for analyzing various activities of the user. The real time applications require analysis of the continuously data generated on mobile devices. Deep learning helps to extract useful information from the huge data being generated. This paper discusses about various deep learning models available for object detection. The idea of measuring relative distances of objects in the image along with object identification is proposed. The state-of-art models only detect the objects in the images, whereas to know the relative position of these objects along with its identification becomes useful and helps to build better vision applications. The paper also describes various methods to run this deep learning model on mobile devices.
Keywords: Deep Neural Network (DNN), Convolution Neural Network(CNN), Deep Learning, Object Detection, MaskRCNN.
Scope of the Article: Mobile App Security and Privacy