Qualcomm RB5 AI Hardware Acceleration with TensorFlow

From RidgeRun Developer Wiki


Follow Us in Twitter LinkedIn Email Share this page



Index






In the section AI Acceleration we had an overview of our options to work Machine Learning/AI applications in the Qualcomm Robotics RB5/RB6. In this section, we are seeing the TensorFlow option. This option comes from an Android specific API that has been ported to the Qualcomm Robotics RB5/RB6 board. We are going to see an example on how to use the GStreamer plugin for TFLite's models and the performance it has in the board.

  • The Example pipeline section shows a functional GStreamer pipeline that used a TFLite model in the Qualcomm Robotics RB5/RB6.
  • The TensorFlow to DLC Conversion section shows how to use tools for converting machine learning models into a format optimized for execution on devices powered by Qualcomm Snapdragon processors.


Index