TDA4VM: Edge-AI Video classification

ravi rk

Part Number: TDA4VM

Hi Team,

We are currently engaged in AI image classification using the SK-TDA4VM platform, and it has been performing according to our expectations. However, we are now considering the extension of our AI capabilities to support video image classification. Does Edge-AI technology in SK-TDA4VM also provide support for video image classification?

Regards,

Raveen

over 1 year ago

0 Praveen Rao over 1 year ago

TI__Mastermind 47943 points

Hello,

The topic expert looking into this topic is currently out-of-office and is expected to be back in 2-3 days. So please expect delay in response to this query.

Thanks.

0 Fabiana Jaimes over 1 year ago

TI__Mastermind 19680 points

Hi Raveen,

Yes. Video and camera stream are supported as inputs with image classification models. We also support object detection which both classifies and localizes and tracks the object with a bounding box. I recommend taking a look at the custom option on the Edge AI Gallery that appears on screen upon boot to see the possibilities for yourself: https://software-dl.ti.com/jacinto7/esd/processor-sdk-linux-edgeai/TDA4VM/09_00_01/exports/docs/common/sample_apps.html

Hope this answered your question.

Thank you,

Fabiana Jaimes

0 ravi rk over 1 year ago in reply to Fabiana Jaimes

Intellectual 925 points

Hello Fabiana Jaimes,

Thank you for the replay, I think I did not explain my question properly, so here is my question,

Model Composer and Model Maker are capable of using "images" as input data to create a trained AI model. However, I would like to confirm if these tools also support "video" as input for training AI models. Additionally, I am interested in knowing whether the TDA4VM or AM62A platform can accommodate and execute these trained models effectively when provided with video input data.

Regards,

Raveen

0 Fabiana Jaimes over 1 year ago in reply to ravi rk

TI__Mastermind 19680 points

Hi Raveen,

I understand your question now. Currently, we only support datasets in JPG and PNG format and do not have any plans to allow for video input. I recommend referring to our model composer quick start guide for additional information: TI Edge AI Studio - Model Composer - Quick Start Guide

Thank you,

Fabiana Jaimes

Processors

Processors forum

TDA4VM: Edge-AI Video classification