ResNet 10 model details used in Nvidia Deepstream

Accelerated Computing Intelligent Video Analytics DeepStream SDK
,
1

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU): Jetson
• DeepStream Version : Any
• JetPack Version (valid for Jetson only): Any
• TensorRT Version : Any
• NVIDIA GPU Driver Version (valid for GPU only): Any
• Issue Type( questions, new requirements, bugs): Any
• How to reproduce the issue ? (This is for bugs. Including which sample app is using, the configuration files content, the command line used and other details for reproducing): Not a bug, so no need to reproduce
• Requirement details( This is for new requirement. Including the module name-for which plugin or for which sample application, the function description): NA

Hi,
I saw that Nvidia Deepstream uses ResNet 10 as an “object detection model”. I am trying to reproduce the results using a custom pipeline but with YOLO.

I would like to know, more details about ResNet 10 that is being used in deepstream. Details like,

  • Model Architecture
  • Number of parameters
  • Output tensor dimension
  • Etc

Is there any document or webpage that describes all these? I tried searching, but didn’t get any information.

Thanks & Regards,
Aravind

3

You can search that from the ngc.

4

Thanks you for the reply @yuweiw

But I didn’t find ResNet 10 details, but it contains details for ResNet 18.

AI Models - Computer Vision, Conversational AI, and More | NVIDIA NGC

Regards,
Aravind

6

The Deepstream caffe model is trained using detectnet_v2 with Resnet10 network and internal data. But the training was not performed using TLT/TAO because TLT/TAO is designed later.
Refer to Dataset used for training sample models.