Actions
Engagement
  • Author Author: rscasny
  • Date Created: Date Created
  • Last Updated Last Updated:
  • Views 374 views
  • Likes 0 likes
  • Comments 3 comments
Related
Recommended

Poll: Seeking Your Interest in Sound Classification and Vision Object Detection WorkShop

I was speaking with a supplier who may be interested in conducting an AI workshop on Sound Classification and Vision Object Detection. This workshop sparked my interest because it combines two senses, which I found unique. But you tell me. I have pasted below the working draft of this workshop. Id like to know if it captures your interest and to what degree. If you have any other ideas or "tweaks" to the proposal, feel free to post a comment below. Thanks.

 

Title: Does Your Application Need Better Eyes and Ears?

 

Description:

Today’s products are becoming more sophisticated as they better understand the world around them. The technology making this possible is AI leveraging neural networks. Today, these networks are being used for a wider variety of tasks and do not require cloud resources. Xilinx MPSoCs make this possible by more efficiently processing these networks at the edge while offering standard Linux software (Ubuntu) and popular AI framework environments (Keras) that AI and embedded developers are familiar with.

 

Product teams are also discovering that the use of these networks, that process different types of sensor data such as microphones and cameras, can take on more sophisticated tasks and produce more reliable results. This workshop brings together a confluence of vision and sound AI models that when used together enable more intelligent products making better application decisions. More specifically, this workshop will take you through the network development flow for sound classification (detecting certain sounds) and vision object detection and walking you through our reference design that has sound and vision sensors sharing meta data that enable a higher level application to leverage certain events it sees and hears.

 

What you will learn:

• What makes the Xilinx MPSoC unique regarding neural network processing at the edge

• How to use Aaware Sonus AI to tune sound classification models (how to retrain models with additional environmental background noise)

• How to use Aaware sound classification models together with Xilinx Model Zoo vision models

• How to access the Aaware sound and Xilinx vision model APIs to build higher level AI based applications at the edge

 

Target markets:

• Security & Surveillance

• Robotics

• Conferencing

• Kiosks

  • in reply to aawarejoe

    The dog barking example sounds pretty cool.  Vision to confirm if its a dog, vicious dog or friendly dog, and if a dog on a leash.

    loose vicious dog RUN !    :-) 

  • Hello there,

     

    My name is Joe Gianelli and I'm the CEO of Aaware Inc. We are in discussion with Avnet about offering a compelling workshop that exposes how the Xilinx MPSoC can be used at the edge to process AI sound and vision models, enabling more sophisticated applications. In this workshop we'll walk you through a reference design that can do one of the following:

     

    1. Detect sneeze….then face detect…then detect is person wearing mask or not (using sound+vision to confirm health concern)
    2. Detect dog barking….then vision to confirm dog or not (using sound+vision to confirm correct identity)

     

    Which of these reference designs would be most compelling use case?

  • I was thinking how the human body utilises these two senses so well.

    • In amongst even a noisy audio scenario we can often still extract our name being called as a trigger to look around, see a familiar face or someone looking at us.
    • Another scenario could be we hear something that sounds dangerous and also alerts us to look around or run - a car skidding, a tree creaking in the wind as it snaps....
    • I hear my refuse lorry coming along, and it cues me to look out a few minutes later to work out where my bin is along the road

     

    In electronics the two systems could also work together to create systems that are more aware of the environment they operate in. To be able to slew a limited field of view camera around in azimuth to where something of interest was heard would be invaluable IMO.