How to deploy the image enhancement algorithm with deepstream

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

How to deploy the image enhancement algorithm with deepstream? The output of nvinfer is the enhanced image. How to return to the pipeline? When I try to sink, I still save the original low brightness image

3

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU)
• DeepStream Version
• 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)

4

• Hardware Platform (Jetson / GPU) ：T4 GPU
• DeepStream Version ：5.1
• JetPack Version (valid for Jetson only)：None
• TensorRT Version：TRT7.1
• NVIDIA GPU Driver Version (valid for GPU only)：450.80.02
• Issue Type( questions, new requirements, bugs) ：Question
• 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)：

Question:I use deepstream to infer an image enhancement model. Its output is an enhanced image, which has been converted into an engine file. I want to realize the output (sink) of the enhanced video. However, I saved MP4 through sink and found that the result MP4 video was not enhanced. What’s the reason? thank you！！！

5

Yeah, I think you need to save the output from the engine. the output sink always saved the original stream.
Such as if your model is a detector, the model just output the tensor containing bbox, DS will not save the tensor from model, it just parse the tensor and attach the bbox to meta data, is that clear for you?

6

Yes, but I want to achieve an effect similar to replacing the original stream. What should I do?

7

I think you need to modify all the corresponding meta data if you want to modify the output buffer, so we don’t suggest you to do that, if you just want to get the output from the model, you can attach the model’s output as tensor meta, you can refer infer tensor meta test app for how to do that

9

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.