OpenVINO Deep Learning Workbench: A Platform for Model Optimization, Analysis and Deployment

Demidovskij, Alexander; Tugaryov, Artyom; Suvorov, A. K.; Tarkan, Yaroslav; Fatekhov, Marat; Сальников, И. И.; Kashchikhin, Andrey; Golubenko, Vladimir; Dedyukhina, Galina; Alborova, Alina; Palmer, Ryan; Fedorov, Mikhail; Gorbachev, Yury

doi:10.1109/ictai50040.2020.00106

Cited by 12 publications

(3 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this regard, it is necessary to convert the weights of the DNN since they are trained in FP32. The conversion is achieved starting with the TensorFlow trained files using a desktop PC and performing post-training quantization with OpenVINO toolkit libraries [36] which are specific for Intel devices. These libraries convert the data of the DNN to FP16 data type and adapt the DNN operations to the NCS architecture.…”

Section: E Deep Learning Neural Accelerators For the Edgementioning

confidence: 99%

Characterizing Deep Neural Networks on Edge Computing Systems for Object Classification in 3D Point Clouds

Wisultschew¹,

Pérez²,

Otero³

et al. 2022

IEEE Sensors J.

View full text Add to dashboard Cite

The current trend of shifting computing from the cloud to the edge of the Internet of Things is influencing deep learning applications. Moving intelligence closer to the point of need entails advantages in terms of performance, power consumption, security, and privacy. The problem arises with data sources that generate a massive amount of information, making data processing challenging for edge devices. This is the case of point clouds generated by LIDAR sensors. Implementations at the edge become even more challenging when heavy processing algorithms such as deep neural networks are selected. However, deep neural networks are the stateof-the-art solution to carry out object classification tasks as they provide the best results in terms of accuracy when working with high data volumes. This work demonstrates that the processing of point cloud-based sensors using deep neural networks at the edge is becoming feasible with the emergence of new devices with high computing capacity combined with reduced power consumption. In this regard, a characterization of first-in-class deep learning classification algorithms working with point cloud data as inputs and running over different state-of-the-art edge processing architectures is provided. A broad range of devices, including CPUs, GPU-based, SoC FPGA-based, and deep learning neural accelerators, have been evaluated in terms of inference time, classification accuracy, and power consumption. As a result, it demonstrates that neural accelerators with integrated host CPUs represent the best trade-off between power consumption and performance, making them a perfect solution for IoT applications at the edge level.

show abstract

Section: E Deep Learning Neural Accelerators For the Edgementioning

confidence: 99%

Characterizing Deep Neural Networks on Edge Computing Systems for Object Classification in 3D Point Clouds

Wisultschew¹,

Pérez²,

Otero³

et al. 2022

IEEE Sensors J.

View full text Add to dashboard Cite

show abstract

“…How to enable fast inference on low-powered embedded platforms remains an open research question. Intel OpenVINO toolkit emerges as an extremely useful tool of choice since it optimizes DL models across Intel hardware while minimizing the inference time [11]. A large portion of the studies discussed above quite commonly neglect this design aspect and demonstrates their DL solutions based on expensive GPU resources.…”

Section: Inference Optimizationmentioning

confidence: 99%

“…An allied question is: How much accuracy and AP we need to sacrifice while pursuing faster inference? In contrast, our measurement outputs are based on OpenVINO DL Workbench [11], which is an open-source and productionready framework to ensure reusability, interoperability, and scalability.…”

Section: Inference Optimizationmentioning

confidence: 99%

An Optimized Multi-Task Learning Model for Disaster Classification and Victim Detection in Federated Learning Environments

et al. 2022

View full text Add to dashboard Cite

show abstract

Machine Learning

2023

Design for Embedded Image Processing on FPGAs

View full text Add to dashboard Cite

OpenVINO Deep Learning Workbench: A Platform for Model Optimization, Analysis and Deployment

Cited by 12 publications

References 12 publications

Characterizing Deep Neural Networks on Edge Computing Systems for Object Classification in 3D Point Clouds

Characterizing Deep Neural Networks on Edge Computing Systems for Object Classification in 3D Point Clouds

An Optimized Multi-Task Learning Model for Disaster Classification and Victim Detection in Federated Learning Environments

Machine Learning

Contact Info

Product

Resources

About