DevMesh

Emotion Detection on Movidius NCS with OpenVino and Intel-optimized TensorFlow

Justin Shenk

Justin Shenk

Unknown

6 0
  • 0 Collaborators

Real-time emotion detection at the edge using a model optimized for edge devices with OpenVino. ...learn more

Project status: Published/In Market

Internet of Things, Artificial Intelligence

Intel Technologies
OpenVINO, Intel Opt ML/DL Framework, Movidius NCS

Code Samples [1]

Overview / Usage

Emotion detection with deep learning at the edge benefits from neuromorphic computing ability of Movidius NCS. The TensorFlow model is further optimized for Intel hardware (Up-squared) using OpenVino and a special TensorFlow build.

Methodology / Approach

  • Train ResNet (deep neural network) on FER2013 facial expression database
  • Compile model for Movidius NCS with OpenVino
  • Run live inference with UpSquared board and webcam.

Technologies Used

OpenVino
UpSquared board
Movidius Neural Compute Stick
Intel-optimized TensorFlow

Repository

https://github.com/justinshenk/party-pi

Comments (0)