Au-Zone Technologies: eIQ Marketplace – How to Use DeepView Middleware to Accelerate Your Intelligent Vision Applications on i.MX 8M Family of Devices | NXP Semiconductors

For best experience this site requires Javascript to be enabled. To enable on your browser, follow our accessibility instructions.

Design Center
Training
Au-Zone Technologies: eIQ Marketplace – How to Use DeepView Middleware to Accelerate Your Intelligent Vision Applications on i.MX 8M Family of Devices

Au-Zone Technologies: eIQ Marketplace – How to Use DeepView Middleware to Accelerate Your Intelligent Vision Applications on i.MX 8M Family of Devices

About This Training

The DeepView Detection ModelPack for Application Processors provides a collection of user trainable, documented models fully optimized for advanced detection tasks on the NPU, which allow you to achieve state-of-the-art precision and performance right out of the box.  The DeepView ModelPack integrates seamlessly into eIQ Portal and extends the examples that come free in the baseline toolkit from NXP.  

With the customized detection model in hand, we will then go through the steps of deploying the model and evaluating the runtime performance, system loading and model accuracy on an i.MX 8M Plus target with DeepView VisionPack. The VisionPack helps customers reduce cost, project risk and development time by providing all the needed GStreamer plugins implemented and pretested as part of a production-grade vision pipeline that uses the available hardware acceleration of the NXP i.MX 8M family of application processors.

Presented by

Sébastien Taylor
VP of R&D, Au-Zone Technologies

DURATION

14 min

LANGUAGE

English

SKILL LEVEL

Beginner

0Views1Likes

We are looking for ways to improve.

Take the survey.

Additional Resources

AI and Machine Learning

i.MX Applications Processors

Have a question? Ask the community.

Ask our i.MX Processors community and get expert advice.

Stay In the Loop

Subscribe to our newsletter to stay updated with our latest developments and if you need further assistance, we are here to help.

Subscribe Contact Support