• State Resolved
  • Locked Locked
  • Replies 2 replies
  • Subscribers 98 subscribers
  • Views 263 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

SK-AM62A-LP: Driver Monitoring System (DMS) Custom Model

Anusha Subramanyam
Prodigy 80 points
Part Number: SK-AM62A-LP


Tool/software:

Hi,

I am currently working on implementing a Driver Monitoring System using the TI AM62A Starter Kit.

The use case is to build a object detection model to detect and alert the driver in case of following conditions.
             1. Eye drowsiness detection
             2. Yawning detection
             3. Distracted detection
             4. Phone call detection
             5. Smoking detection
             6. No driver detection
             7. Head phone detection to the extent possible
             8. Camera covered

I wanted to inquire if there is any pre-trained custom model available for DMS consisting of any of the above features that we could use directly for our project. If such a model exists, could you please guide me on how to access and implement it for our use case?
Looking forward for your support and assistance.
Regards,
Anusha Subramanyam
  • 0
    TI__Genius 16136 points

    Hello Anusha,

    We do not have pre-trained models for these tasks, currently. You will need to develop models and procure a good dataset that encompasses these tasks. 

    Our offering for models is generally not application-specific, like for DMS / OMS. We have an array of models within our modelzoo that can give you an idea of which architectures are supported. Developers typically use multiple models with application-specific heads for this. You may use any of the supported classifier / feature extraction networks (e.g. mobilenet, resnet) as a starting point for training here. 

    BR,
    Reese

  • 0 in reply to Reese Grimsley
    Prodigy 80 points

    Hello Reese,

    Thank You