Deploying Deep Learning Models for Various Real-Time Applications Using Keras

Deep learning solutions are being increasingly used in mobile applications. Although there are many open-source software tools for the development of deep learning solutions, there are no guidelines in one place in a unified manner for using these tools toward real-time deployment of these solutions on smartphones. From the variety of available deep learning tools, the most suited ones are used in this paper to enable real-time deployment of deep learning inference networks on smartphones. A uniform flow of implementation is devised for both Android and iOS smartphones. The advantage of using multi-threading to achieve or improve real-time throughputs is also showcased. A benchmarking framework consisting of accuracy, CPU/GPU consumption, and real-time throughput is considered for validation purposes. The developed deployment approach allows deep learning models to be turned into real-time smartphone apps with ease based on publicly available deep learning and smartphone software tools. This approach is applied to six popular or representative convolutional neural network models, and the validation results based on the benchmarking metrics are reported.

Download Full-text

Sa2032 PIXEL VS PATCH-BASED DEEP LEARNING MODELS, PAVING THE WAY TOWARDS REAL-TIME COMPUTER-ASSISSTED DETECTION OF BARRETT’S NEOPLASIA

Gastrointestinal Endoscopy ◽

10.1016/j.gie.2020.03.1828 ◽

2020 ◽

Vol 91 (6) ◽

pp. AB251

Author(s):

Mohamed Abdelrahim ◽

Masahiro Saiko ◽

Yukiko Masaike ◽

E.J.A.Z. HOSSAIN ◽

Sophie Arndtz ◽

...

Keyword(s):

Deep Learning ◽

Real Time ◽

Learning Models ◽

The Way

Download Full-text

Real-Time Vehicle Make and Model Recognition with the Residual SqueezeNet Architecture

Sensors ◽

10.3390/s19050982 ◽

2019 ◽

Vol 19 (5) ◽

pp. 982 ◽

Cited By ~ 9

Author(s):

Hyo Lee ◽

Ihsan Ullah ◽

Weiguo Wan ◽

Yongbin Gao ◽

Zhijun Fang

Keyword(s):

Deep Learning ◽

Real Time ◽

Large Scale ◽

Recognition Rate ◽

Experimental Results ◽

Learning Approach ◽

Deep Model ◽

Proposed Model ◽

Real Time Applications ◽

Model Recognition

Make and model recognition (MMR) of vehicles plays an important role in automatic vision-based systems. This paper proposes a novel deep learning approach for MMR using the SqueezeNet architecture. The frontal views of vehicle images are first extracted and fed into a deep network for training and testing. The SqueezeNet architecture with bypass connections between the Fire modules, a variant of the vanilla SqueezeNet, is employed for this study, which makes our MMR system more efficient. The experimental results on our collected large-scale vehicle datasets indicate that the proposed model achieves 96.3% recognition rate at the rank-1 level with an economical time slice of 108.8 ms. For inference tasks, the deployed deep model requires less than 5 MB of space and thus has a great viability in real-time applications.

Download Full-text

Deep Learning Models for Real-time Human Activity Recognition with Smartphones

Mobile Networks and Applications ◽

10.1007/s11036-019-01445-x ◽

2019 ◽

Vol 25 (2) ◽

pp. 743-755 ◽

Cited By ~ 18

Author(s):

Shaohua Wan ◽

Lianyong Qi ◽

Xiaolong Xu ◽

Chao Tong ◽

Zonghua Gu

Keyword(s):

Deep Learning ◽

Real Time ◽

Activity Recognition ◽

Human Activity ◽

Human Activity Recognition ◽

Learning Models

Download Full-text

The Development and Application of Real-Time Deep Learning Models to Drive Directional Drilling Efficiency

10.2118/199584-ms ◽

2020 ◽

Cited By ~ 1

Author(s):

Dingzhou Cao ◽

Don Hender ◽

Sam Ariabod ◽

Chris James ◽

Yuxing Ben ◽

...

Keyword(s):

Deep Learning ◽

Real Time ◽

Directional Drilling ◽

Learning Models

Download Full-text

Ransomware Traffic Classification Using Deep Learning Models

International Journal of Web Portals ◽

10.4018/ijwp.2020010101 ◽

2020 ◽

Vol 12 (1) ◽

pp. 1-11

Author(s):

Arivudainambi D. ◽

Varun Kumar K.A. ◽

Vinoth Kumar R. ◽

Visu P.

Keyword(s):

Deep Learning ◽

Real Time ◽

Network Traffic ◽

Classification Model ◽

Traffic Classification ◽

Learning Models ◽

Learning Methods ◽

Novel Method

Ransomware is a malware which affects the systems data with modern encryption techniques, and the data is recovered once a ransom amount is paid. In this research, the authors show how ransomware propagates and infects devices. Live traffic classifications of ransomware have been meticulously analyzed. Further, a novel method for the classification of ransomware traffic by using deep learning methods is presented. Based on classification, the detection of ransomware is approached with the characteristics of the network traffic and its communications. In more detail, the behavior of popular ransomware, Crypto Wall, is analyzed and based on this knowledge, a real-time ransomware live traffic classification model is proposed.

Download Full-text