Lane detection using deepstream

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU)- Jetson AGX Xavier
• DeepStream Version-6.1
• JetPack Version (valid for Jetson only)
• TensorRT Version
• NVIDIA GPU Driver Version (valid for GPU only)
• Issue Type( questions, new requirements, bugs)
• 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)
• Requirement details( This is for new requirement. Including the module name-for which plugin or for which sample application, the function description)

Hi, I want to perform lane detection using deepstream. I did not find any Lane detection models in the deepstream (Please let me If there is any model available in the deepstream). so I want to implement the ultra-fast-lane-detection-v2 model, available here at GitHub - cfzd/Ultra-Fast-Lane-Detection-v2: Ultra Fast Deep Lane Detection With Hybrid Anchor Driven Ordinal Classification (TPAMI 2022). I do not understand how to run this model on deepstream. I want to use the pretrained weights given in the page at GitHub - cfzd/Ultra-Fast-Lane-Detection-v2: Ultra Fast Deep Lane Detection With Hybrid Anchor Driven Ordinal Classification (TPAMI 2022) and run the model in deepstream using these weights. From documentation here Using a Custom Model with DeepStream — DeepStream 6.2 Release documentation I understand that I have to modify the nvinfer plugin, but I dont understand how can I get all these files and link them in deepstream.

Also I am confused about using TAO toolkit. will using TAO toolkit help my case?

Please let me know the best way and steps involved to do this.

3

There is no update from you for a period, assuming this is not an issue anymore. Hence we are closing this topic. If need further support, please open a new one. Thanks

No. Currently there is no model provided by DeepStream. All sampe models are from TAO.

Seems this is a Pytorch model. It is better to use ONNX model than to use Pytorch model with DeepStream. Pytorch has provided method to convert to ONNX model. torch.onnx — PyTorch 1.13 documentation

What do you mean by “all these files”? Which files?

There is TAO forum. Latest Intelligent Video Analytics/TAO Toolkit topics - NVIDIA Developer Forums

5

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.